Hugging Face's logo

jmainz
/
wav2vec2-mms-1b-l1107-mus-asr-10m

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
jmainz commited on
Commit
ceeda07
•
1 Parent(s): 5c4d3e2

End of training

Browse files
Files changed (7) hide show
  1. README.md +6 -14
  2. adapter.mus.safetensors +1 -1
  3. model.safetensors +1 -1
  4. runs/May19_23-17-25_patas-gn2/events.out.tfevents.1716188026.patas-gn2.1330444.1 +3 -0
  5. runs/May20_00-15-13_patas-gn3/events.out.tfevents.1716189322.patas-gn3.737810.0 +3 -0
  6. runs/May20_01-27-59_patas-gn3/events.out.tfevents.1716193687.patas-gn3.738442.0 +3 -0
  7. training_args.bin +1 -1
README.md CHANGED
@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [facebook/mms-1b-l1107](https://huggingface.co/facebook/mms-1b-l1107) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 2.8676
21
- - Wer: 1.0
22
- - Cer: 0.8667
23
 
24
  ## Model description
25
 
@@ -42,25 +42,17 @@ The following hyperparameters were used during training:
42
  - train_batch_size: 8
43
  - eval_batch_size: 8
44
  - seed: 42
45
- - gradient_accumulation_steps: 4
46
- - total_train_batch_size: 32
47
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
  - lr_scheduler_type: linear
49
  - lr_scheduler_warmup_steps: 100
50
- - num_epochs: 8
 
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
55
  |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
56
- | 22.7436 | 0.8889 | 6 | 27.6744 | 1.9219 | 3.2356 |
57
- | 17.6284 | 1.9259 | 13 | 23.4047 | 2.1406 | 1.6089 |
58
- | 13.6566 | 2.9630 | 20 | 15.5898 | 1.0312 | 0.9089 |
59
- | 7.9631 | 4.0 | 27 | 6.4957 | 1.0 | 0.9311 |
60
- | 4.7531 | 4.8889 | 33 | 3.6537 | 1.0 | 1.0 |
61
- | 3.1476 | 5.9259 | 40 | 3.4255 | 1.0 | 1.0 |
62
- | 2.8137 | 6.9630 | 47 | 2.9391 | 1.0 | 0.9289 |
63
- | 1.9846 | 7.1111 | 48 | 2.8676 | 1.0 | 0.8667 |
64
 
65
 
66
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [facebook/mms-1b-l1107](https://huggingface.co/facebook/mms-1b-l1107) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 2.4292
21
+ - Wer: 0.7031
22
+ - Cer: 0.1778
23
 
24
  ## Model description
25
 
 
42
  - train_batch_size: 8
43
  - eval_batch_size: 8
44
  - seed: 42
 
 
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_steps: 100
48
+ - num_epochs: 4
49
+ - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
54
  |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
55
+ | 6.5188 | 3.7037 | 100 | 2.4292 | 0.7031 | 0.1778 |
 
 
 
 
 
 
 
56
 
57
 
58
  ### Framework versions
adapter.mus.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0cf563cc58ff1ad27b436a5905f3dbf4abca6fe2ea8d76e7df2a5db8072c9174
3
  size 8834400
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b81946aadc188c5bd02d581eb8bf2daebe0f6160775c2eae02154c52c0f96922
3
  size 8834400
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8622d3317a14229777b12467c9a80cfa80f2126d2cfc1d0da053764e4eb1b956
3
  size 3858926736
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cb178a2d1d847202ebe05c8b0d04dd909d89851a0fd335cdfeb6463a4de2bcb
3
  size 3858926736
runs/May19_23-17-25_patas-gn2/events.out.tfevents.1716188026.patas-gn2.1330444.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7af8590acba3f5e47207f0975c5c67f9e56287a6c173059d192af691182c894f
3
+ size 446
runs/May20_00-15-13_patas-gn3/events.out.tfevents.1716189322.patas-gn3.737810.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41e642e5107395589d4ea58de692d8ebdfffff28c79a48508a125d1dc64e2419
3
+ size 6680
runs/May20_01-27-59_patas-gn3/events.out.tfevents.1716193687.patas-gn3.738442.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c9e0a98379f0ecd9655a63aea4d7aa5a020eff69011b126b7161376b0c00bad
3
+ size 7028
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:efd15792fe60474f7ed7ad912a781bb5a437b41d2cf55fee828a0bd83632fa53
3
  size 4527
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c18323b1c85697ccfd6dd986c4bea42d730d63d92373661caf88c4e5cf234a3
3
  size 4527