Upload 81 files

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/49epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13942d8df6c2326cc0f4f5df1745ba9a6ee792b54ca65b85c322e9ebdad81439
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/50epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a034944163cbf78cb911dd3e9e7e5561b8516429ce774c7669808a12a8f459b
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/53epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc4a2014dacb245d48ec2e038cb352ff0902f18bc0156ce2b94eb1d321f2a5ab
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/54epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50c499c23cef8489938aeb927367d479b55b109e852fde28fa17a14af64a8a76
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/55epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2930d5d0bb9b0029c8906d01748174b5663905ab969ee2137f019f2a72a02c7c
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/56epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d295b116b9891ef8520e7987d0682f79636305c96362e7067cfa63096f9e2e2e
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/57epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec090b65d90f017b160f5718d059cbbec01424491ae31ddb62e16255f64cc80a
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/58epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07d4fb13b7ec8fcbc0e79b48c2f2e3ca03db84a62c6aecb96db6ec70d992daee
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/59epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63fced0fdc8817cd4f376fe8e33ffc3b07a8615ce2151f6b5bd6be0990bb75ff
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/60epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ab99d72231e8b471e6f66d902a88c36315286ecfc4bb77cf1efcd73bfe200db
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/RESULTS.md

@@ -0,0 +1,196 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Tue Mar  5 09:47:19 CST 2024`
+- python version: `3.9.18 (main, Sep 11 2023, 13:41:44)  [GCC 11.2.0]`
+- espnet version: `espnet 202308`
+- pytorch version: `pytorch 1.12.1+cu116`
+- Git hash: `884659f9ee95374811015381c976fa3b4f6e01db`
+  - Commit date: `Thu Nov 23 00:23:29 2023 +0800`
+
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_30epoch/test_clean_kaldi_fmt|961|64007|80.7|15.1|4.2|6.7|26.0|98.9|
+|decode_sot_asr_model_30epoch/test_other_kaldi_fmt|992|80370|75.0|19.6|5.4|7.8|32.9|99.6|
+|decode_sot_asr_model_35epoch/test_clean_kaldi_fmt|961|64007|81.5|14.5|4.0|7.4|26.0|98.6|
+|decode_sot_asr_model_35epoch/test_other_kaldi_fmt|992|80370|75.7|18.8|5.5|9.0|33.2|99.4|
+|decode_sot_asr_model_40epoch/test_clean_kaldi_fmt|961|64007|80.4|14.2|5.4|4.8|24.4|97.7|
+|decode_sot_asr_model_40epoch/test_other_kaldi_fmt|992|80370|74.8|18.1|7.1|5.7|30.9|99.5|
+|decode_sot_asr_model_8epoch/test_clean_kaldi_fmt|961|64007|70.4|23.7|5.8|7.2|36.7|99.1|
+|decode_sot_asr_model_8epoch/test_other_kaldi_fmt|992|80370|63.7|28.7|7.5|8.6|44.8|99.6|
+|decode_sot_asr_model_valid.acc.best/dev|3000|126853|54.3|32.2|13.4|28.9|74.6|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|226216|76.8|11.5|11.6|12.4|35.5|99.2|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_sys7_8khz_spk1|1606|135101|36.6|21.7|41.7|3.2|66.7|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_sys7_8khz_spk2|1606|135101|35.5|21.5|43.0|3.4|67.9|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|209679|63.9|20.3|15.8|10.4|46.5|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk_kaldi_fmt|1004|124462|67.6|17.1|15.2|7.2|39.6|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|200029|52.0|27.4|20.7|11.8|59.9|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk_kaldi_fmt|721|119166|55.5|22.9|21.6|11.3|55.8|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_oracle|544|10798|85.9|12.2|1.9|88.7|102.9|92.8|
+|decode_sot_asr_model_valid.acc.best/eval_oracle|4479|96585|84.7|13.0|2.4|88.2|103.6|94.5|
+|decode_sot_asr_model_valid.acc.best/sot_sdm1_dev|2382|35243|0.0|0.0|100.0|0.0|100.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|301042|77.5|10.9|11.6|11.3|33.8|99.5|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|212871|64.4|19.3|16.3|11.8|47.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|185394|53.2|26.2|20.6|10.6|57.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|336490|75.9|13.3|10.8|11.6|35.6|99.9|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|266074|60.5|23.6|15.9|11.5|51.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|259138|49.2|30.2|20.6|11.1|61.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test|3000|114243|55.6|30.7|13.7|32.9|77.3|99.9|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|178761|80.2|9.0|10.8|7.5|27.2|98.9|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_sys7_8khz_spk1|2180|178761|33.1|21.4|45.4|3.0|69.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_sys7_8khz_spk2|2180|178761|31.6|24.4|44.0|3.0|71.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_3spk_kaldi_fmt|977|124741|66.8|17.5|15.7|9.0|42.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_4spk_kaldi_fmt|632|109072|56.3|22.4|21.3|8.8|52.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_kaldi_fmt|961|64007|81.4|13.3|5.3|5.9|24.5|97.5|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|205496|78.6|11.7|9.6|7.8|29.1|99.8|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_sys7_8khz_spk1|2363|205496|28.2|26.2|45.6|3.1|74.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_sys7_8khz_spk2|2363|205496|35.6|22.3|42.1|3.1|67.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_3spk_kaldi_fmt|1246|162996|62.5|21.9|15.6|9.1|46.6|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_4spk_kaldi_fmt|901|157123|51.9|26.0|22.1|10.4|58.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_kaldi_fmt|992|80370|75.5|17.6|6.9|6.8|31.3|99.3|
+|decode_sot_css_asr_model_valid.acc.best/dev_oracle|544|10798|85.0|12.9|2.1|48.0|63.0|92.8|
+|decode_sot_css_asr_model_valid.acc.best/eval_oracle|4479|96585|84.1|13.2|2.7|49.1|65.0|94.5|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_30epoch/test_clean_kaldi_fmt|961|329390|89.3|6.0|4.6|6.7|17.4|98.9|
+|decode_sot_asr_model_30epoch/test_other_kaldi_fmt|992|416899|85.6|8.2|6.3|7.7|22.2|99.6|
+|decode_sot_asr_model_35epoch/test_clean_kaldi_fmt|961|329390|89.8|5.8|4.4|7.4|17.6|98.6|
+|decode_sot_asr_model_35epoch/test_other_kaldi_fmt|992|416899|85.9|8.0|6.1|8.5|22.6|99.4|
+|decode_sot_asr_model_40epoch/test_clean_kaldi_fmt|961|329390|88.6|5.9|5.5|4.5|16.0|97.7|
+|decode_sot_asr_model_40epoch/test_other_kaldi_fmt|992|416899|84.5|8.0|7.5|5.2|20.7|99.5|
+|decode_sot_asr_model_8epoch/test_clean_kaldi_fmt|961|329390|83.8|9.3|6.9|7.2|23.5|99.1|
+|decode_sot_asr_model_8epoch/test_other_kaldi_fmt|992|416899|78.7|12.1|9.2|8.2|29.5|99.6|
+|decode_sot_asr_model_valid.acc.best/dev|3000|673222|71.1|13.3|15.6|28.5|57.5|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|1230801|83.7|6.2|10.1|10.8|27.2|99.2|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_sys7_8khz_spk1|1606|735694|47.9|9.1|43.0|3.6|55.7|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_sys7_8khz_spk2|1606|735694|47.1|9.1|43.8|3.7|56.5|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|1140428|74.6|10.3|15.2|9.3|34.7|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk_kaldi_fmt|1004|677017|76.8|8.2|15.0|6.2|29.4|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|1087409|65.5|13.1|21.3|10.3|44.7|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk_kaldi_fmt|721|647884|67.3|10.5|22.2|8.9|41.6|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_oracle|544|57590|94.8|3.2|2.0|86.2|91.5|92.8|
+|decode_sot_asr_model_valid.acc.best/eval_oracle|4479|522239|93.8|3.6|2.6|85.2|91.4|94.5|
+|decode_sot_asr_model_valid.acc.best/sot_sdm1_dev|2382|169857|0.0|0.0|100.0|0.0|100.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|1550429|84.0|6.1|10.0|9.9|25.9|99.5|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|1084475|74.6|10.4|15.1|10.2|35.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|938467|66.3|13.1|20.7|9.9|43.6|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|1742136|83.5|7.0|9.4|10.0|26.5|99.9|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|1381987|72.6|12.0|15.4|10.0|37.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|1346646|63.9|14.4|21.7|9.9|46.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test|3000|608408|71.7|12.5|15.8|32.4|60.7|99.9|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|921344|85.5|5.0|9.5|6.1|20.6|98.9|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_sys7_8khz_spk1|2180|921344|44.6|8.9|46.5|3.5|58.8|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_sys7_8khz_spk2|2180|921344|44.9|10.2|44.9|3.7|58.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_3spk_kaldi_fmt|977|635802|76.0|9.0|15.0|7.7|31.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_4spk_kaldi_fmt|632|552325|67.8|10.6|21.6|7.8|39.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_kaldi_fmt|961|329390|89.2|5.3|5.4|5.8|16.6|97.5|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|1064868|85.3|6.1|8.6|6.4|21.1|99.8|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_sys7_8khz_spk1|2363|1064868|42.2|11.1|46.7|3.8|61.6|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_sys7_8khz_spk2|2363|1064868|47.8|9.2|43.0|3.7|55.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_3spk_kaldi_fmt|1246|847159|73.8|10.5|15.6|7.9|34.1|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_4spk_kaldi_fmt|901|817228|65.0|11.9|23.1|8.7|43.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_kaldi_fmt|992|416899|85.1|7.3|7.6|6.8|21.7|99.3|
+|decode_sot_css_asr_model_valid.acc.best/dev_oracle|544|57590|93.7|4.0|2.4|46.5|52.8|92.8|
+|decode_sot_css_asr_model_valid.acc.best/eval_oracle|4479|522239|92.9|4.1|3.0|47.2|54.3|94.5|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/decode_sot_asr_model_30epoch
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|47659|78.4|16.5|5.1|5.5|27.1|97.2|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|258151|87.5|6.5|6.1|5.6|18.1|97.2|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/decode_sot_asr_model_35epoch
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|47659|79.4|16.2|4.4|6.8|27.4|98.5|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|258151|88.2|6.5|5.3|6.7|18.5|98.5|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/decode_sot_asr_model_40epoch
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|47659|78.5|15.6|5.8|4.0|25.5|97.5|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|258151|87.0|6.3|6.7|3.9|16.9|97.5|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/decode_sot_asr_model_8epoch
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|47659|68.4|24.9|6.7|6.7|38.2|98.3|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|258151|81.6|9.9|8.5|6.7|25.2|98.3|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/decode_sot_asr_model_valid.acc.best
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_2spk_kaldi_fmt|1606|135101|79.1|9.8|11.1|8.4|29.2|98.4|
+|org/dev_kaldi_fmt|605|47659|78.8|15.0|6.2|5.7|26.9|97.5|
+|org/sot_sdm1_eval|2385|37529|33.4|53.9|12.7|102.0|168.6|100.0|
+|org/tt_mix_clean_reverb_max_16k|3000|3000|0.0|100.0|0.0|4047.2|4147.2|100.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_2spk_kaldi_fmt|1606|735694|84.8|5.3|9.9|6.9|22.1|98.4|
+|org/dev_kaldi_fmt|605|258151|87.1|6.0|6.9|5.7|18.6|97.5|
+|org/sot_sdm1_eval|2385|183036|62.5|22.7|14.8|97.8|135.2|100.0|
+|org/tt_mix_clean_reverb_max_16k|3000|143026|17.3|82.6|0.1|411.1|493.8|100.0|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/checkpoint.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14eb244b2d0dfc6d2dc23b8b9a10e1814df51f163d680c62439ff31374e6bd2e
+size 516820547

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/config.yaml

@@ -0,0 +1,227 @@
+config: conf/tuning/train_sot_asr_conformer_medium.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new
+ngpu: 1
+seed: 0
+num_workers: 16
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 4
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 44319
+dist_launcher: null
+multiprocessing_distributed: true
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 60
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - acc
+    - max
+keep_nbest_models: 10
+nbest_averaging_interval: 0
+grad_clip: 5.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 4
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_matplotlib: true
+use_tensorboard: true
+create_graph_in_tensorboard: false
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param:
+- /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 20
+valid_batch_size: null
+batch_bins: 16000000
+valid_batch_bins: null
+train_shape_file:
+- exp/asr_stats_raw_en_char_sp/train/speech_shape
+- exp/asr_stats_raw_en_char_sp/train/text_shape.char
+valid_shape_file:
+- exp/asr_stats_raw_en_char_sp/valid/speech_shape
+- exp/asr_stats_raw_en_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+train_data_path_and_name_and_type:
+-   - dump/raw/train_medium_kaldi_fmt_sp/wav.scp
+    - speech
+    - kaldi_ark
+-   - dump/raw/train_medium_kaldi_fmt_sp/text
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev_kaldi_fmt/wav.scp
+    - speech
+    - kaldi_ark
+-   - dump/raw/dev_kaldi_fmt/text
+    - text
+    - text
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+exclude_weight_decay: false
+exclude_weight_decay_conf: {}
+optim: adam
+optim_conf:
+    lr: 0.002
+    weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 20000
+token_list:
+- <blank>
+- <unk>
+- <sc>
+- <space>
+- E
+- T
+- A
+- O
+- N
+- I
+- H
+- S
+- R
+- D
+- L
+- U
+- M
+- C
+- W
+- F
+- G
+- Y
+- P
+- B
+- V
+- K
+- ''''
+- X
+- J
+- Q
+- Z
+- <sos/eos>
+init: null
+input_size: null
+ctc_conf:
+    dropout_rate: 0.0
+    ctc_type: builtin
+    reduce: true
+    ignore_nan_grad: null
+    zero_infinity: true
+joint_net_conf: null
+use_preprocessor: true
+token_type: char
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: null
+speech_volume_normalize: null
+rir_scp: null
+rir_apply_prob: 1.0
+noise_scp: null
+noise_apply_prob: 1.0
+noise_db_range: '13_15'
+short_noise_thres: 0.5
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+    fs: 16k
+specaug: null
+specaug_conf: {}
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/asr_stats_raw_en_char_sp/train/feats_stats.npz
+model: espnet
+model_conf:
+    ctc_weight: 0.0
+    lsm_weight: 0.1
+    length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: conformer
+encoder_conf:
+    output_size: 256
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 12
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.1
+    input_layer: conv2d
+    normalize_before: true
+    macaron_style: true
+    rel_pos_type: latest
+    pos_enc_layer_type: rel_pos
+    selfattention_layer_type: rel_selfattn
+    activation_type: swish
+    use_cnn_module: true
+    cnn_module_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 6
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.1
+    src_attention_dropout_rate: 0.1
+preprocessor: multi
+preprocessor_conf:
+    speaker_change_symbol:
+    - <sc>
+required:
+- output_dir
+- token_list
+version: '202308'
+distributed: true

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/latest.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ab99d72231e8b471e6f66d902a88c36315286ecfc4bb77cf1efcd73bfe200db
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/run.sh

@@ -0,0 +1 @@
+./asr.sh --lang en --audio_format flac.ark --stage 12 --feats_type raw --token_type char --sot_asr true --max_wav_duration 50 --speed_perturb_factors '0.9 1.0 1.1' --feats_normalize global_mvn --use_lm false --pretrained_model /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth --asr_config conf/tuning/train_sot_asr_conformer_medium.yaml --lm_config conf/tuning/train_lm_transformer.yaml --inference_config conf/tuning/decode_sot.yaml --train_set train_medium_kaldi_fmt --valid_set dev_kaldi_fmt --test_sets 'dev_kaldi_fmt test_clean_kaldi_fmt test_other_kaldi_fmt' --ngpu 4 --asr_tag train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new --lm_train_text data/local/other_text/text --bpe_train_text data/train_medium_kaldi_fmt/text --stage 12 "$@"; exit $?

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/train/events.out.tfevents.1700307214.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.1753232.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c30cf181e3c82c4cc12d7aebfb77b819aa2da1fd08a2d567519ad1b5196e3171
+size 88

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/train/events.out.tfevents.1700310367.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.2045481.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0848df7dac3f565c375a3f38dc7a661fd41628f61bffb3d75e70e3641706faec
+size 107387014

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/train/events.out.tfevents.1700412292.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.1823743.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c40b67a8b2a99f63326ec1b7b8685ca88cc3ed776a10e39c5273e164c8982b67
+size 8306

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/train/events.out.tfevents.1700536345.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.2992215.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08432dbcf98ebaad27388a1b90b781f1aef124b1ac2c3c4676f8a387c81c0c1a
+size 743299213

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/valid/events.out.tfevents.1700307214.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.1753232.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05d778de3f4cc8bdfe77155d9f1e2d6782358f94fdcf930fd16cb4b391e0f81f
+size 88

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/valid/events.out.tfevents.1700310367.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.2045481.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ead35607a81bfff4e5bab9f6dd496480516f7350927ad98a7a4db115507b2ee
+size 2338

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/valid/events.out.tfevents.1700412292.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.1823743.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e422f849535779325246177b90cb5193453b58ac6cceb2e15640105036c361a8
+size 88

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/valid/events.out.tfevents.1700536345.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.2992215.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:143371316cfd5edd14fe5c0cc235d84980859e4dd329af2b61fcdd64eb7b93fb
+size 14752

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/train.1.log

@@ -0,0 +1,773 @@
+# python3 -m espnet2.bin.asr_train --use_preprocessor true --bpemodel none --token_type char --token_list data/en_token_list/char/tokens.txt --non_linguistic_symbols none --cleaner none --g2p none --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/wav.scp,speech,kaldi_ark --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/speech_shape --resume true --init_param /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth --ignore_init_mismatch false --fold_length 80000 --output_dir exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new --config conf/tuning/train_sot_asr_conformer_medium.yaml --frontend_conf fs=16k --normalize=global_mvn --normalize_conf stats_file=exp/asr_stats_raw_en_char_sp/train/feats_stats.npz --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/wav.scp,speech,kaldi_ark --train_shape_file exp/asr_stats_raw_en_char_sp/train/speech_shape --fold_length 150 --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/text,text,text --train_shape_file exp/asr_stats_raw_en_char_sp/train/text_shape.char --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/text,text,text --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/text_shape.char --ngpu 4 --multiprocessing_distributed True 
+# Started at Mon Nov 20 00:43:12 CST 2023
+#
+/star-home/jinzengrui/lib/miniconda3/envs/dev39/bin/python3 /star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py --use_preprocessor true --bpemodel none --token_type char --token_list data/en_token_list/char/tokens.txt --non_linguistic_symbols none --cleaner none --g2p none --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/wav.scp,speech,kaldi_ark --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/speech_shape --resume true --init_param /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth --ignore_init_mismatch false --fold_length 80000 --output_dir exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new --config conf/tuning/train_sot_asr_conformer_medium.yaml --frontend_conf fs=16k --normalize=global_mvn --normalize_conf stats_file=exp/asr_stats_raw_en_char_sp/train/feats_stats.npz --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/wav.scp,speech,kaldi_ark --train_shape_file exp/asr_stats_raw_en_char_sp/train/speech_shape --fold_length 150 --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/text,text,text --train_shape_file exp/asr_stats_raw_en_char_sp/train/text_shape.char --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/text,text,text --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/text_shape.char --ngpu 4 --multiprocessing_distributed True
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:23,119 (distributed_c10d:228) INFO: Added key: store_based_barrier_key:1 to store for rank: 0
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:23,119 (distributed_c10d:262) INFO: Rank 0: Completed store-based barrier for key:store_based_barrier_key:1 with 4 nodes.
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:23,174 (asr:490) INFO: Vocabulary size: 32
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:31,610 (abs_task:1229) INFO: pytorch.version=1.11.0+cu102, cuda.available=True, cudnn.version=7605, cudnn.benchmark=False, cudnn.deterministic=True
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:31,624 (abs_task:1230) INFO: Model structure:
+ESPnetASRModel(
+  (frontend): DefaultFrontend(
+    (stft): Stft(n_fft=512, win_length=512, hop_length=128, center=True, normalized=False, onesided=True)
+    (frontend): Frontend()
+    (logmel): LogMel(sr=16000, n_fft=512, n_mels=80, fmin=0, fmax=8000.0, htk=False)
+  )
+  (normalize): GlobalMVN(stats_file=exp/asr_stats_raw_en_char_sp/train/feats_stats.npz, norm_means=True, norm_vars=True)
+  (encoder): ConformerEncoder(
+    (embed): Conv2dSubsampling(
+      (conv): Sequential(
+        (0): Conv2d(1, 256, kernel_size=(3, 3), stride=(2, 2))
+        (1): ReLU()
+        (2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2))
+        (3): ReLU()
+      )
+      (out): Sequential(
+        (0): Linear(in_features=4864, out_features=256, bias=True)
+        (1): RelPositionalEncoding(
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+      )
+    )
+    (encoders): MultiSequential(
+      (0): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (1): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (2): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (3): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (4): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (5): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (6): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (7): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (8): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (9): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (10): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (11): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+    )
+    (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+  )
+  (decoder): TransformerDecoder(
+    (embed): Sequential(
+      (0): Embedding(32, 256)
+      (1): PositionalEncoding(
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+    )
+    (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+    (output_layer): Linear(in_features=256, out_features=32, bias=True)
+    (decoders): MultiSequential(
+      (0): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (1): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (2): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (3): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (4): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (5): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+    )
+  )
+  (criterion_att): LabelSmoothingLoss(
+    (criterion): KLDivLoss()
+  )
+)
+
+Model summary:
+    Class Name: ESPnetASRModel
+    Total Number of model parameters: 43.00 M
+    Number of trainable parameters: 43.00 M (100.0%)
+    Size: 172.01 MB
+    Type: torch.float32
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:31,624 (abs_task:1233) INFO: Optimizer:
+Adam (
+Parameter Group 0
+    amsgrad: False
+    betas: (0.9, 0.999)
+    eps: 1e-08
+    initial_lr: 0.002
+    lr: 1e-07
+    maximize: False
+    weight_decay: 1e-06
+)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:31,624 (abs_task:1234) INFO: Scheduler: WarmupLR(warmup_steps=20000)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:31,626 (abs_task:1243) INFO: Saving the configuration in exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/config.yaml
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:31,692 (abs_task:1304) INFO: Loading pretrained params from /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:35,293 (asr:461) INFO: Optional Data Names: ('text_spk2', 'text_spk3', 'text_spk4')
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,615 (abs_task:1614) INFO: [train] dataset:
+ESPnetDataset(
+  speech: {"path": "dump/raw/train_medium_kaldi_fmt_sp/wav.scp", "type": "kaldi_ark"}
+  text: {"path": "dump/raw/train_medium_kaldi_fmt_sp/text", "type": "text"}
+  preprocess: <espnet2.train.preprocessor.CommonPreprocessor_multi object at 0x7f4603b63730>)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,616 (abs_task:1615) INFO: [train] Batch sampler: NumElementsBatchSampler(N-batch=10615, batch_bins=16000000, sort_in_batch=descending, sort_batch=descending)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,620 (abs_task:1616) INFO: [train] mini-batch sizes summary: N-batch=10615, mean=53.4, min=7, max=201
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,737 (asr:461) INFO: Optional Data Names: ('text_spk2', 'text_spk3', 'text_spk4')
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,747 (abs_task:1614) INFO: [valid] dataset:
+ESPnetDataset(
+  speech: {"path": "dump/raw/dev_kaldi_fmt/wav.scp", "type": "kaldi_ark"}
+  text: {"path": "dump/raw/dev_kaldi_fmt/text", "type": "text"}
+  preprocess: <espnet2.train.preprocessor.CommonPreprocessor_multi object at 0x7f45f1f856d0>)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,747 (abs_task:1615) INFO: [valid] Batch sampler: NumElementsBatchSampler(N-batch=12, batch_bins=16000000, sort_in_batch=descending, sort_batch=descending)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,747 (abs_task:1616) INFO: [valid] mini-batch sizes summary: N-batch=12, mean=50.4, min=17, max=82
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,752 (asr:461) INFO: Optional Data Names: ('text_spk2', 'text_spk3', 'text_spk4')
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,775 (abs_task:1614) INFO: [plot_att] dataset:
+ESPnetDataset(
+  speech: {"path": "dump/raw/dev_kaldi_fmt/wav.scp", "type": "kaldi_ark"}
+  text: {"path": "dump/raw/dev_kaldi_fmt/text", "type": "text"}
+  preprocess: <espnet2.train.preprocessor.CommonPreprocessor_multi object at 0x7f45f1f85d60>)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,775 (abs_task:1615) INFO: [plot_att] Batch sampler: UnsortedBatchSampler(N-batch=605, batch_size=1, key_file=exp/asr_stats_raw_en_char_sp/valid/speech_shape, 
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:44,775 (abs_task:1616) INFO: [plot_att] mini-batch sizes summary: N-batch=3, mean=1.0, min=1, max=1
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:48,183 (trainer:159) INFO: The training was resumed using exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/checkpoint.pth
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823743 [0] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823743 [0] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823743 [0] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823743 [0] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823743 [0] NCCL INFO Using network Socket
+NCCL version 2.10.3+cuda10.2
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823744 [1] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823745 [2] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823746 [3] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823744 [1] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823745 [2] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823746 [3] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823744 [1] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823744 [1] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823744 [1] NCCL INFO Using network Socket
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823746 [3] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823746 [3] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823746 [3] NCCL INFO Using network Socket
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823745 [2] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823745 [2] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823745 [2] NCCL INFO Using network Socket
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Trees [0] -1/-1/-1->3->2 [1] -1/-1/-1->3->2
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Setting affinity for GPU 3 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Trees [0] 3/-1/-1->2->1 [1] 3/-1/-1->2->1
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Setting affinity for GPU 2 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Channel 00/02 :    0   1   2   3
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Trees [0] 2/-1/-1->1->0 [1] 2/-1/-1->1->0
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Channel 01/02 :    0   1   2   3
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Setting affinity for GPU 1 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Setting affinity for GPU 0 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Channel 00 : 3[8000] -> 0[4000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Channel 00 : 2[7000] -> 3[8000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Channel 01 : 2[7000] -> 3[8000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Channel 00 : 1[6000] -> 2[7000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Channel 01 : 1[6000] -> 2[7000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Channel 00 : 0[4000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Channel 01 : 0[4000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Channel 01 : 3[8000] -> 0[4000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Channel 00 : 3[8000] -> 2[7000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Channel 01 : 3[8000] -> 2[7000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Channel 00 : 2[7000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Channel 00 : 1[6000] -> 0[4000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Channel 01 : 2[7000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Channel 01 : 1[6000] -> 0[4000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823746:1823875 [3] NCCL INFO comm 0x7f4294001200 rank 3 nranks 4 cudaDev 3 busId 8000 - Init COMPLETE
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823870 [0] NCCL INFO comm 0x7f450c001200 rank 0 nranks 4 cudaDev 0 busId 4000 - Init COMPLETE
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823745:1823876 [2] NCCL INFO comm 0x7f631c001200 rank 2 nranks 4 cudaDev 2 busId 7000 - Init COMPLETE
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823744:1823874 [1] NCCL INFO comm 0x7fa7ac001200 rank 1 nranks 4 cudaDev 1 busId 6000 - Init COMPLETE
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1823743:1823743 [0] NCCL INFO Launch mode Parallel
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:44:52,106 (trainer:284) INFO: 9/60epoch started
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:49:07,241 (distributed:948) INFO: Reducer buckets have been rebuilt in this iteration.
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 00:55:25,795 (trainer:732) INFO: 9epoch:train:1-530batch: iter_time=0.002, forward_time=0.209, loss_att=107.878, acc=0.904, loss=107.878, backward_time=0.325, grad_norm=103.272, clip=100.000, loss_scale=1.000, optim_step_time=0.085, optim0_lr0=0.002, train_time=4.789
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:01:45,507 (trainer:732) INFO: 9epoch:train:531-1060batch: iter_time=2.446e-04, forward_time=0.209, loss_att=112.672, acc=0.903, loss=112.672, backward_time=0.327, grad_norm=105.518, clip=100.000, loss_scale=1.000, optim_step_time=0.084, optim0_lr0=0.002, train_time=2.865
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:08:04,727 (trainer:732) INFO: 9epoch:train:1061-1590batch: iter_time=2.303e-04, forward_time=0.208, loss_att=109.250, acc=0.905, loss=109.250, backward_time=0.327, grad_norm=112.951, clip=100.000, loss_scale=1.000, optim_step_time=0.084, optim0_lr0=0.002, train_time=2.863
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:14:24,051 (trainer:732) INFO: 9epoch:train:1591-2120batch: iter_time=3.562e-04, forward_time=0.211, loss_att=109.274, acc=0.906, loss=109.274, backward_time=0.327, grad_norm=104.180, clip=100.000, loss_scale=1.000, optim_step_time=0.086, optim0_lr0=0.002, train_time=2.861
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:20:44,930 (trainer:732) INFO: 9epoch:train:2121-2650batch: iter_time=3.394e-04, forward_time=0.211, loss_att=110.221, acc=0.905, loss=110.221, backward_time=0.328, grad_norm=110.200, clip=100.000, loss_scale=1.000, optim_step_time=0.086, optim0_lr0=0.002, train_time=2.875
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:27:04,841 (trainer:732) INFO: 9epoch:train:2651-3180batch: iter_time=3.402e-04, forward_time=0.211, loss_att=111.572, acc=0.905, loss=111.572, backward_time=0.328, grad_norm=102.797, clip=100.000, loss_scale=1.000, optim_step_time=0.085, optim0_lr0=0.002, train_time=2.866
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:33:23,833 (trainer:732) INFO: 9epoch:train:3181-3710batch: iter_time=3.449e-04, forward_time=0.210, loss_att=108.396, acc=0.906, loss=108.396, backward_time=0.327, grad_norm=103.265, clip=100.000, loss_scale=1.000, optim_step_time=0.085, optim0_lr0=0.002, train_time=2.860
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:39:42,244 (trainer:732) INFO: 9epoch:train:3711-4240batch: iter_time=3.845e-04, forward_time=0.210, loss_att=105.457, acc=0.907, loss=105.457, backward_time=0.326, grad_norm=105.516, clip=100.000, loss_scale=1.000, optim_step_time=0.086, optim0_lr0=0.002, train_time=2.855
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:46:01,633 (trainer:732) INFO: 9epoch:train:4241-4770batch: iter_time=3.412e-04, forward_time=0.211, loss_att=110.350, acc=0.906, loss=110.350, backward_time=0.327, grad_norm=104.195, clip=100.000, loss_scale=1.000, optim_step_time=0.083, optim0_lr0=0.002, train_time=2.863
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:52:20,391 (trainer:732) INFO: 9epoch:train:4771-5300batch: iter_time=3.219e-04, forward_time=0.210, loss_att=106.497, acc=0.907, loss=106.497, backward_time=0.327, grad_norm=105.330, clip=100.000, loss_scale=1.000, optim_step_time=0.084, optim0_lr0=0.002, train_time=2.858
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 01:58:41,240 (trainer:732) INFO: 9epoch:train:5301-5830batch: iter_time=3.283e-04, forward_time=0.211, loss_att=108.151, acc=0.909, loss=108.151, backward_time=0.328, grad_norm=100.440, clip=100.000, loss_scale=1.000, optim_step_time=0.084, optim0_lr0=0.002, train_time=2.875
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 02:05:01,053 (trainer:732) INFO: 9epoch:train:5831-6360batch: iter_time=3.129e-04, forward_time=0.210, loss_att=107.337, acc=0.908, loss=107.337, backward_time=0.328, grad_norm=105.336, clip=100.000, loss_scale=1.000, optim_step_time=0.083, optim0_lr0=0.002, train_time=2.865
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 02:11:21,060 (trainer:732) INFO: 9epoch:train:6361-6890batch: iter_time=3.236e-04, forward_time=0.210, loss_att=106.133, acc=0.909, loss=106.133, backward_time=0.328, grad_norm=104.171, clip=100.000, loss_scale=1.000, optim_step_time=0.083, optim0_lr0=0.002, train_time=2.869
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-20 02:17:40,657 (trainer:732) INFO: 9epoch:train:6891-7420batch: iter_time=3.511e-04, forward_time=0.210, loss_att=107.243, acc=0.908, loss=107.243, backward_time=0.327, grad_norm=108.454, clip=100.000, loss_scale=1.000, optim_step_time=0.085, optim0_lr0=0.002, train_time=2.863
+Traceback (most recent call last):
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 197, in _run_module_as_main
+    return _run_code(code, main_globals, None,
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 87, in _run_code
+    exec(code, run_globals)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 23, in <module>
+    main()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 19, in main
+    ASRTask.main(cmd=cmd)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 1132, in main
+    while not ProcessContext(processes, error_queues).join():
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/multiprocessing/spawn.py", line 140, in join
+    raise ProcessExitedException(
+torch.multiprocessing.spawn.ProcessExitedException: process 0 terminated with signal SIGKILL
+# Accounting: time=5952 threads=1
+# Ended (code 1) at Mon Nov 20 02:22:24 CST 2023, elapsed time 5952 seconds
+/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/resource_tracker.py:216: UserWarning: resource_tracker: There appear to be 224 leaked semaphore objects to clean up at shutdown
+  warnings.warn('resource_tracker: There appear to be %d '

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/train.3.log

@@ -0,0 +1,1057 @@
+# python3 -m espnet2.bin.asr_train --use_preprocessor true --bpemodel none --token_type char --token_list data/en_token_list/char/tokens.txt --non_linguistic_symbols none --cleaner none --g2p none --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/wav.scp,speech,kaldi_ark --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/speech_shape --resume true --init_param /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth --ignore_init_mismatch false --fold_length 80000 --output_dir exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new --config conf/tuning/train_sot_asr_conformer_medium.yaml --frontend_conf fs=16k --normalize=global_mvn --normalize_conf stats_file=exp/asr_stats_raw_en_char_sp/train/feats_stats.npz --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/wav.scp,speech,kaldi_ark --train_shape_file exp/asr_stats_raw_en_char_sp/train/speech_shape --fold_length 150 --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/text,text,text --train_shape_file exp/asr_stats_raw_en_char_sp/train/text_shape.char --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/text,text,text --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/text_shape.char --ngpu 4 --multiprocessing_distributed True 
+# Started at Sat Nov 18 19:32:54 CST 2023
+#
+/star-home/jinzengrui/lib/miniconda3/envs/dev39/bin/python3 /star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py --use_preprocessor true --bpemodel none --token_type char --token_list data/en_token_list/char/tokens.txt --non_linguistic_symbols none --cleaner none --g2p none --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/wav.scp,speech,kaldi_ark --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/speech_shape --resume true --init_param /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth --ignore_init_mismatch false --fold_length 80000 --output_dir exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new --config conf/tuning/train_sot_asr_conformer_medium.yaml --frontend_conf fs=16k --normalize=global_mvn --normalize_conf stats_file=exp/asr_stats_raw_en_char_sp/train/feats_stats.npz --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/wav.scp,speech,kaldi_ark --train_shape_file exp/asr_stats_raw_en_char_sp/train/speech_shape --fold_length 150 --train_data_path_and_name_and_type dump/raw/train_medium_kaldi_fmt_sp/text,text,text --train_shape_file exp/asr_stats_raw_en_char_sp/train/text_shape.char --valid_data_path_and_name_and_type dump/raw/dev_kaldi_fmt/text,text,text --valid_shape_file exp/asr_stats_raw_en_char_sp/valid/text_shape.char --ngpu 4 --multiprocessing_distributed True
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:14,486 (distributed_c10d:228) INFO: Added key: store_based_barrier_key:1 to store for rank: 0
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:14,486 (distributed_c10d:262) INFO: Rank 0: Completed store-based barrier for key:store_based_barrier_key:1 with 4 nodes.
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:14,530 (asr:490) INFO: Vocabulary size: 32
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:22,927 (abs_task:1229) INFO: pytorch.version=1.11.0+cu102, cuda.available=True, cudnn.version=7605, cudnn.benchmark=False, cudnn.deterministic=True
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:22,941 (abs_task:1230) INFO: Model structure:
+ESPnetASRModel(
+  (frontend): DefaultFrontend(
+    (stft): Stft(n_fft=512, win_length=512, hop_length=128, center=True, normalized=False, onesided=True)
+    (frontend): Frontend()
+    (logmel): LogMel(sr=16000, n_fft=512, n_mels=80, fmin=0, fmax=8000.0, htk=False)
+  )
+  (normalize): GlobalMVN(stats_file=exp/asr_stats_raw_en_char_sp/train/feats_stats.npz, norm_means=True, norm_vars=True)
+  (encoder): ConformerEncoder(
+    (embed): Conv2dSubsampling(
+      (conv): Sequential(
+        (0): Conv2d(1, 256, kernel_size=(3, 3), stride=(2, 2))
+        (1): ReLU()
+        (2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2))
+        (3): ReLU()
+      )
+      (out): Sequential(
+        (0): Linear(in_features=4864, out_features=256, bias=True)
+        (1): RelPositionalEncoding(
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+      )
+    )
+    (encoders): MultiSequential(
+      (0): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (1): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (2): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (3): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (4): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (5): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (6): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (7): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (8): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (9): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (10): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (11): EncoderLayer(
+        (self_attn): RelPositionMultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (linear_pos): Linear(in_features=256, out_features=256, bias=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (feed_forward_macaron): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): Swish()
+        )
+        (conv_module): ConvolutionModule(
+          (pointwise_conv1): Conv1d(256, 512, kernel_size=(1,), stride=(1,))
+          (depthwise_conv): Conv1d(256, 256, kernel_size=(31,), stride=(1,), padding=(15,), groups=256)
+          (norm): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
+          (pointwise_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+          (activation): Swish()
+        )
+        (norm_ff): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_mha): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_ff_macaron): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_conv): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm_final): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+    )
+    (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+  )
+  (decoder): TransformerDecoder(
+    (embed): Sequential(
+      (0): Embedding(32, 256)
+      (1): PositionalEncoding(
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+    )
+    (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+    (output_layer): Linear(in_features=256, out_features=32, bias=True)
+    (decoders): MultiSequential(
+      (0): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (1): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (2): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (3): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (4): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+      (5): DecoderLayer(
+        (self_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (src_attn): MultiHeadedAttention(
+          (linear_q): Linear(in_features=256, out_features=256, bias=True)
+          (linear_k): Linear(in_features=256, out_features=256, bias=True)
+          (linear_v): Linear(in_features=256, out_features=256, bias=True)
+          (linear_out): Linear(in_features=256, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+        )
+        (feed_forward): PositionwiseFeedForward(
+          (w_1): Linear(in_features=256, out_features=2048, bias=True)
+          (w_2): Linear(in_features=2048, out_features=256, bias=True)
+          (dropout): Dropout(p=0.1, inplace=False)
+          (activation): ReLU()
+        )
+        (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (norm3): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+        (dropout): Dropout(p=0.1, inplace=False)
+      )
+    )
+  )
+  (criterion_att): LabelSmoothingLoss(
+    (criterion): KLDivLoss()
+  )
+)
+
+Model summary:
+    Class Name: ESPnetASRModel
+    Total Number of model parameters: 43.00 M
+    Number of trainable parameters: 43.00 M (100.0%)
+    Size: 172.01 MB
+    Type: torch.float32
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:22,942 (abs_task:1233) INFO: Optimizer:
+Adam (
+Parameter Group 0
+    amsgrad: False
+    betas: (0.9, 0.999)
+    eps: 1e-08
+    initial_lr: 0.002
+    lr: 1e-07
+    maximize: False
+    weight_decay: 1e-06
+)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:22,942 (abs_task:1234) INFO: Scheduler: WarmupLR(warmup_steps=20000)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:22,945 (abs_task:1243) INFO: Saving the configuration in exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/config.yaml
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:22,967 (abs_task:1304) INFO: Loading pretrained params from /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:25,928 (asr:461) INFO: Optional Data Names: ('text_spk2', 'text_spk3', 'text_spk4')
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,548 (abs_task:1614) INFO: [train] dataset:
+ESPnetDataset(
+  speech: {"path": "dump/raw/train_medium_kaldi_fmt_sp/wav.scp", "type": "kaldi_ark"}
+  text: {"path": "dump/raw/train_medium_kaldi_fmt_sp/text", "type": "text"}
+  preprocess: <espnet2.train.preprocessor.CommonPreprocessor_multi object at 0x7f8e24d29490>)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,548 (abs_task:1615) INFO: [train] Batch sampler: NumElementsBatchSampler(N-batch=9484, batch_bins=16000000, sort_in_batch=descending, sort_batch=descending)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,552 (abs_task:1616) INFO: [train] mini-batch sizes summary: N-batch=9484, mean=41.8, min=10, max=170
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,657 (asr:461) INFO: Optional Data Names: ('text_spk2', 'text_spk3', 'text_spk4')
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,665 (abs_task:1614) INFO: [valid] dataset:
+ESPnetDataset(
+  speech: {"path": "dump/raw/dev_kaldi_fmt/wav.scp", "type": "kaldi_ark"}
+  text: {"path": "dump/raw/dev_kaldi_fmt/text", "type": "text"}
+  preprocess: <espnet2.train.preprocessor.CommonPreprocessor_multi object at 0x7f8e12793700>)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,665 (abs_task:1615) INFO: [valid] Batch sampler: NumElementsBatchSampler(N-batch=11, batch_bins=16000000, sort_in_batch=descending, sort_batch=descending)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,665 (abs_task:1616) INFO: [valid] mini-batch sizes summary: N-batch=11, mean=41.1, min=22, max=66
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,671 (asr:461) INFO: Optional Data Names: ('text_spk2', 'text_spk3', 'text_spk4')
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,697 (abs_task:1614) INFO: [plot_att] dataset:
+ESPnetDataset(
+  speech: {"path": "dump/raw/dev_kaldi_fmt/wav.scp", "type": "kaldi_ark"}
+  text: {"path": "dump/raw/dev_kaldi_fmt/text", "type": "text"}
+  preprocess: <espnet2.train.preprocessor.CommonPreprocessor_multi object at 0x7f8e12793c70>)
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,697 (abs_task:1615) INFO: [plot_att] Batch sampler: UnsortedBatchSampler(N-batch=452, batch_size=1, key_file=exp/asr_stats_raw_en_char_sp/valid/speech_shape, 
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:32,697 (abs_task:1616) INFO: [plot_att] mini-batch sizes summary: N-batch=3, mean=1.0, min=1, max=1
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1753232 [0] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1753232 [0] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1753232 [0] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1753232 [0] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1753232 [0] NCCL INFO Using network Socket
+NCCL version 2.10.3+cuda10.2
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1753233 [1] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1753234 [2] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1753233 [1] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1753234 [2] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1753235 [3] NCCL INFO Bootstrap : Using eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1753235 [3] NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1753233 [1] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1753233 [1] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1753233 [1] NCCL INFO Using network Socket
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1753234 [2] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1753234 [2] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1753234 [2] NCCL INFO Using network Socket
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1753235 [3] NCCL INFO NET/IB : No device found.
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1753235 [3] NCCL INFO NET/Socket : Using [0]eth0:10.177.13.150<0>
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1753235 [3] NCCL INFO Using network Socket
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Trees [0] 2/-1/-1->1->0 [1] 2/-1/-1->1->0
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Setting affinity for GPU 1 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Trees [0] 3/-1/-1->2->1 [1] 3/-1/-1->2->1
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Trees [0] -1/-1/-1->3->2 [1] -1/-1/-1->3->2
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Channel 00/02 :    0   1   2   3
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Setting affinity for GPU 2 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Setting affinity for GPU 5 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Channel 01/02 :    0   1   2   3
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Trees [0] 1/-1/-1->0->-1 [1] 1/-1/-1->0->-1
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Setting affinity for GPU 0 to 20000002
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Channel 00 : 2[7000] -> 3[e000] via direct shared memory
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Channel 01 : 2[7000] -> 3[e000] via direct shared memory
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Channel 00 : 1[6000] -> 2[7000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Channel 01 : 1[6000] -> 2[7000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Channel 00 : 0[4000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Channel 01 : 0[4000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Channel 00 : 3[e000] -> 0[4000] via direct shared memory
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Channel 01 : 3[e000] -> 0[4000] via direct shared memory
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Channel 00 : 1[6000] -> 0[4000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Channel 01 : 1[6000] -> 0[4000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Connected all rings
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Channel 00 : 3[e000] -> 2[7000] via direct shared memory
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Channel 01 : 3[e000] -> 2[7000] via direct shared memory
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Channel 00 : 2[7000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Channel 01 : 2[7000] -> 1[6000] via P2P/IPC
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO Connected all trees
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO threadThresholds 8/8/64 | 32/8/64 | 8/8/512
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO 2 coll channels, 2 p2p channels, 2 p2p channels per peer
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753233:1756016 [1] NCCL INFO comm 0x7fc0a8001200 rank 1 nranks 4 cudaDev 1 busId 6000 - Init COMPLETE
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753235:1756018 [3] NCCL INFO comm 0x7f483c001200 rank 3 nranks 4 cudaDev 3 busId e000 - Init COMPLETE
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1756014 [0] NCCL INFO comm 0x7f8d5c001200 rank 0 nranks 4 cudaDev 0 busId 4000 - Init COMPLETE
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753232:1753232 [0] NCCL INFO Launch mode Parallel
+de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:1753234:1756017 [2] NCCL INFO comm 0x7fbe0c001200 rank 2 nranks 4 cudaDev 2 busId 7000 - Init COMPLETE
+[de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb:0/4] 2023-11-18 19:33:34,575 (trainer:284) INFO: 1/60epoch started
+Traceback (most recent call last):
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 197, in _run_module_as_main
+    return _run_code(code, main_globals, None,
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 87, in _run_code
+Traceback (most recent call last):
+  File "<string>", line 1, in <module>
+Traceback (most recent call last):
+  File "<string>", line 1, in <module>
+Traceback (most recent call last):
+Traceback (most recent call last):
+  File "<string>", line 1, in <module>
+  File "<string>", line 1, in <module>
+    exec(code, run_globals)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 23, in <module>
+    main()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 19, in main
+    ASRTask.main(cmd=cmd)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 1132, in main
+    while not ProcessContext(processes, error_queues).join():
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/multiprocessing/spawn.py", line 109, in join
+    ready = multiprocessing.connection.wait(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/connection.py", line 931, in wait
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 116, in spawn_main
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 116, in spawn_main
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 116, in spawn_main
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 116, in spawn_main
+    ready = selector.select(timeout)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/selectors.py", line 416, in select
+    exitcode = _main(fd, parent_sentinel)
+    exitcode = _main(fd, parent_sentinel)
+    exitcode = _main(fd, parent_sentinel)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 125, in _main
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 125, in _main
+    exitcode = _main(fd, parent_sentinel)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 125, in _main
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 125, in _main
+    prepare(preparation_data)
+    prepare(preparation_data)
+    prepare(preparation_data)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 234, in prepare
+    prepare(preparation_data)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 234, in prepare
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 234, in prepare
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 234, in prepare
+    fd_event_list = self._selector.poll(timeout)
+KeyboardInterrupt
+    _fixup_main_from_name(data['init_main_from_name'])
+    _fixup_main_from_name(data['init_main_from_name'])
+    _fixup_main_from_name(data['init_main_from_name'])
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 258, in _fixup_main_from_name
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 258, in _fixup_main_from_name
+    _fixup_main_from_name(data['init_main_from_name'])
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 258, in _fixup_main_from_name
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/spawn.py", line 258, in _fixup_main_from_name
+    main_content = runpy.run_module(mod_name,
+    main_content = runpy.run_module(mod_name,
+    main_content = runpy.run_module(mod_name,
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 225, in run_module
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 225, in run_module
+    main_content = runpy.run_module(mod_name,
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 225, in run_module
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 225, in run_module
+    return _run_module_code(code, init_globals, run_name, mod_spec)
+    return _run_module_code(code, init_globals, run_name, mod_spec)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 97, in _run_module_code
+    return _run_module_code(code, init_globals, run_name, mod_spec)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 97, in _run_module_code
+    return _run_module_code(code, init_globals, run_name, mod_spec)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 97, in _run_module_code
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 97, in _run_module_code
+    _run_code(code, mod_globals, init_globals,
+    _run_code(code, mod_globals, init_globals,
+    _run_code(code, mod_globals, init_globals,
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 87, in _run_code
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 87, in _run_code
+    _run_code(code, mod_globals, init_globals,
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 87, in _run_code
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/runpy.py", line 87, in _run_code
+    exec(code, run_globals)
+    exec(code, run_globals)
+    exec(code, run_globals)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 2, in <module>
+    exec(code, run_globals)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 2, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 2, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/bin/asr_train.py", line 2, in <module>
+    from espnet2.tasks.asr import ASRTask
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/asr.py", line 73, in <module>
+    from espnet2.tasks.asr import ASRTask
+    from espnet2.tasks.asr import ASRTask
+    from espnet2.tasks.asr import ASRTask
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/asr.py", line 73, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/asr.py", line 73, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/asr.py", line 73, in <module>
+    from espnet2.tasks.abs_task import AbsTask
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 48, in <module>
+    from espnet2.tasks.abs_task import AbsTask
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 48, in <module>
+    from espnet2.tasks.abs_task import AbsTask
+    from espnet2.tasks.abs_task import AbsTask
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 48, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 48, in <module>
+    from espnet2.train.dataset import DATA_TYPES, AbsDataset, ESPnetDataset
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/dataset.py", line 10, in <module>
+    from espnet2.train.dataset import DATA_TYPES, AbsDataset, ESPnetDataset
+    from espnet2.train.dataset import DATA_TYPES, AbsDataset, ESPnetDataset
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/dataset.py", line 10, in <module>
+    from espnet2.train.dataset import DATA_TYPES, AbsDataset, ESPnetDataset
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/dataset.py", line 10, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/dataset.py", line 10, in <module>
+    import h5py
+    import h5py
+    import h5py
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/__init__.py", line 58, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/__init__.py", line 58, in <module>
+    import h5py
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/__init__.py", line 58, in <module>
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/__init__.py", line 58, in <module>
+    from ._hl import filters
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/_hl/filters.py", line 44, in <module>
+    from ._hl import filters
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/_hl/filters.py", line 44, in <module>
+    from ._hl import filters
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/_hl/filters.py", line 44, in <module>
+    from ._hl import filters
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/h5py/_hl/filters.py", line 44, in <module>
+    from .base import product
+  File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
+    from .base import product
+  File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
+    from .base import product
+  File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
+    from .base import product
+  File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
+  File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
+  File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
+  File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
+  File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
+  File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
+  File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
+  File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
+  File "<frozen importlib._bootstrap_external>", line 846, in exec_module
+  File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
+  File "<frozen importlib._bootstrap_external>", line 846, in exec_module
+  File "<frozen importlib._bootstrap_external>", line 846, in exec_module
+  File "<frozen importlib._bootstrap_external>", line 941, in get_code
+  File "<frozen importlib._bootstrap_external>", line 941, in get_code
+  File "<frozen importlib._bootstrap_external>", line 941, in get_code
+  File "<frozen importlib._bootstrap_external>", line 1039, in get_data
+  File "<frozen importlib._bootstrap_external>", line 1039, in get_data
+  File "<frozen importlib._bootstrap_external>", line 846, in exec_module
+KeyboardInterrupt
+  File "<frozen importlib._bootstrap_external>", line 1039, in get_data
+  File "<frozen importlib._bootstrap_external>", line 941, in get_code
+KeyboardInterrupt
+KeyboardInterrupt
+  File "<frozen importlib._bootstrap_external>", line 1039, in get_data
+KeyboardInterrupt
+Process SpawnProcess-3:
+Process SpawnProcess-1:
+Process SpawnProcess-4:
+Process SpawnProcess-2:
+Traceback (most recent call last):
+Traceback (most recent call last):
+Traceback (most recent call last):
+Traceback (most recent call last):
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
+    self.run()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
+    self.run()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 108, in run
+    self._target(*self._args, **self._kwargs)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
+    self.run()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 108, in run
+    self._target(*self._args, **self._kwargs)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 1391, in main_worker
+    cls.trainer.run(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
+    self.run()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 108, in run
+    self._target(*self._args, **self._kwargs)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 1391, in main_worker
+    cls.trainer.run(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 290, in run
+    all_steps_are_invalid = cls.train_one_epoch(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 108, in run
+    self._target(*self._args, **self._kwargs)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 1391, in main_worker
+    cls.trainer.run(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 290, in run
+    all_steps_are_invalid = cls.train_one_epoch(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 510, in train_one_epoch
+    for iiter, (utt_id, batch) in enumerate(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/tasks/abs_task.py", line 1391, in main_worker
+    cls.trainer.run(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 290, in run
+    all_steps_are_invalid = cls.train_one_epoch(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 510, in train_one_epoch
+    for iiter, (utt_id, batch) in enumerate(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/reporter.py", line 263, in measure_iter_time
+    iterator = iter(iterable)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 290, in run
+    all_steps_are_invalid = cls.train_one_epoch(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 510, in train_one_epoch
+    for iiter, (utt_id, batch) in enumerate(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/reporter.py", line 263, in measure_iter_time
+    iterator = iter(iterable)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 368, in __iter__
+    return self._get_iterator()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/trainer.py", line 510, in train_one_epoch
+    for iiter, (utt_id, batch) in enumerate(
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/reporter.py", line 263, in measure_iter_time
+    iterator = iter(iterable)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 368, in __iter__
+    return self._get_iterator()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 314, in _get_iterator
+    return _MultiProcessingDataLoaderIter(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/espnet-202308-py3.9.egg/espnet2/train/reporter.py", line 263, in measure_iter_time
+    iterator = iter(iterable)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 368, in __iter__
+    return self._get_iterator()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 314, in _get_iterator
+    return _MultiProcessingDataLoaderIter(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 927, in __init__
+    w.start()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 368, in __iter__
+    return self._get_iterator()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 314, in _get_iterator
+    return _MultiProcessingDataLoaderIter(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 927, in __init__
+    w.start()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 121, in start
+    self._popen = self._Popen(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 314, in _get_iterator
+    return _MultiProcessingDataLoaderIter(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 927, in __init__
+    w.start()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 121, in start
+    self._popen = self._Popen(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 224, in _Popen
+    return _default_context.get_context().Process._Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 927, in __init__
+    w.start()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 121, in start
+    self._popen = self._Popen(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 224, in _Popen
+    return _default_context.get_context().Process._Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 224, in _Popen
+    return _default_context.get_context().Process._Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 284, in _Popen
+    return Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/process.py", line 121, in start
+    self._popen = self._Popen(self)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 284, in _Popen
+    return Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 284, in _Popen
+    return Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 32, in __init__
+    super().__init__(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 224, in _Popen
+    return _default_context.get_context().Process._Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 32, in __init__
+    super().__init__(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 32, in __init__
+    super().__init__(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_fork.py", line 19, in __init__
+    self._launch(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_fork.py", line 19, in __init__
+    self._launch(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/context.py", line 284, in _Popen
+    return Popen(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_fork.py", line 19, in __init__
+    self._launch(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 62, in _launch
+    f.write(fp.getbuffer())
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 62, in _launch
+    f.write(fp.getbuffer())
+KeyboardInterrupt
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 32, in __init__
+    super().__init__(process_obj)
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 62, in _launch
+    f.write(fp.getbuffer())
+KeyboardInterrupt
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_fork.py", line 19, in __init__
+    self._launch(process_obj)
+KeyboardInterrupt
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/multiprocessing/popen_spawn_posix.py", line 62, in _launch
+    f.write(fp.getbuffer())
+KeyboardInterrupt
+Exception ignored in: <function _MultiProcessingDataLoaderIter.__del__ at 0x7fc1763ce5e0>
+Traceback (most recent call last):
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1358, in __del__
+Exception ignored in: <function _MultiProcessingDataLoaderIter.__del__ at 0x7f490af825e0>
+Traceback (most recent call last):
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1358, in __del__
+Exception ignored in: <function _MultiProcessingDataLoaderIter.__del__ at 0x7fbedaa735e0>
+Traceback (most recent call last):
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1358, in __del__
+    self._shutdown_workers()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1316, in _shutdown_workers
+    self._shutdown_workers()
+    self._shutdown_workers()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1316, in _shutdown_workers
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1316, in _shutdown_workers
+Exception ignored in: <function _MultiProcessingDataLoaderIter.__del__ at 0x7f8e352eb5e0>
+Traceback (most recent call last):
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1358, in __del__
+    if self._persistent_workers or self._workers_status[worker_id]:
+    if self._persistent_workers or self._workers_status[worker_id]:
+    if self._persistent_workers or self._workers_status[worker_id]:
+AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute '_workers_status'
+AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute '_workers_status'
+AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute '_workers_status'
+    self._shutdown_workers()
+  File "/star-home/jinzengrui/lib/miniconda3/envs/dev39/lib/python3.9/site-packages/torch/utils/data/dataloader.py", line 1316, in _shutdown_workers
+    if self._persistent_workers or self._workers_status[worker_id]:
+AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute '_workers_status'

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/valid.acc.ave.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6942cd8e2a7d9015a74726b4164bd69324059cb230e97ebccf434f6b1afdee4c
+size 172358249

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/valid.acc.ave_10best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6942cd8e2a7d9015a74726b4164bd69324059cb230e97ebccf434f6b1afdee4c
+size 172358249

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/valid.acc.best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63fced0fdc8817cd4f376fe8e33ffc3b07a8615ce2151f6b5bd6be0990bb75ff
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/48epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5a83dbcc31cbe211bf57d554185f00fdefa382c828fadc33a1202749a7cafbc
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/52epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8359149a4f68dfaf1fb54be64f57b199f0c2a2ce86c123ca9479a8f29854ab59
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/53epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f7928e027e6aaca6f547f40d08d36ed21ec877c81f13b6e8432ffd119a4f635
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/54epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2917e31143d7136e849de1dee47a99942384a481fe11ed989b17d7faf676170c
+size 172367337

medium/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/55epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:035c566070194c490d66447e070ae13154064c7d387b0e57a13693deabadd34d
+size 172367337