jonatasgrosman/exp_w2v2r_es_xls-r_age_teens-2_sixties-8_s182

.gitattributes

@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+cv7_transcriptions.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,17 @@
+---
+language:
+- es
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- es
+datasets:
+- mozilla-foundation/common_voice_7_0
+---
+# exp_w2v2r_es_xls-r_age_teens-2_sixties-8_s182
+
+Fine-tuned [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) for speech recognition using the train split of [Common Voice 7.0 (es)](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0).
+When using this model, make sure that your speech input is sampled at 16kHz.
+
+This model has been fine-tuned by the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) tool.
+    

config.json

@@ -0,0 +1,108 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "activation_dropout": 0.05,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.05,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 768,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.05,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.05,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.05,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.05,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "num_negatives": 100,
+  "output_hidden_size": 1024,
+  "pad_token_id": 0,
+  "proj_codevector_dim": 768,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.18.0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 41,
+  "xvector_output_dim": 512
+}

cv7_evaluation.json

@@ -0,0 +1,300 @@
+{
+  "validation": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.04151456888335528,
+      "wer": 0.1512034750920926
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.0414687375999093,
+      "wer": 0.15101700821936437
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.04347879593393881,
+      "wer": 0.15790634418878188
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.03615933912069448,
+      "wer": 0.13284365162644282
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.04019153605710612,
+      "wer": 0.152372142196379
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.039941026672027875,
+      "wer": 0.14606741573033707
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.053099970335212104,
+      "wer": 0.1582943925233645
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.03064516129032258,
+      "wer": 0.1320754716981132
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.04176513142030384,
+      "wer": 0.15526164462334674
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.04141022740128974,
+      "wer": 0.15089010830569857
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.051328502415458936,
+      "wer": 0.19710669077757687
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.04288797153762904,
+      "wer": 0.15424851795885156
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.03753150966296331,
+      "wer": 0.14190595089471494
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.04139508669562046,
+      "wer": 0.1510940083958784
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.043183836589698044,
+      "wer": 0.14801202806548613
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.03984578802779509,
+      "wer": 0.14775397373876986
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.04328018223234624,
+      "wer": 0.16091954022988506
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.05218601414820828,
+      "wer": 0.1802604523646333
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.03521643433602348,
+      "wer": 0.14220183486238533
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.04713290716534682,
+      "wer": 0.1628175519630485
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.03565980167810832,
+      "wer": 0.13669467787114845
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.040294859063478725,
+      "wer": 0.14418921992267456
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.03947993350748041,
+      "wer": 0.15365329512893983
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.05012410921611018,
+      "wer": 0.16447985004686036
+    }
+  ],
+  "test": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.04834854171536436,
+      "wer": 0.16602910960035414
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.048541338664235996,
+      "wer": 0.166239913097455
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.049413651398215894,
+      "wer": 0.1751859851211903
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.04362688544203735,
+      "wer": 0.14749034749034748
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.042619422773913375,
+      "wer": 0.1560919540229885
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.04365890205694047,
+      "wer": 0.14331853048041987
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.049553050913330746,
+      "wer": 0.15570934256055363
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.10602910602910603,
+      "wer": 0.2926829268292683
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.05486284289276808,
+      "wer": 0.19291819291819293
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.020134228187919462,
+      "wer": 0.18518518518518517
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.048600672359708566,
+      "wer": 0.16642267921459278
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.07747489239598278,
+      "wer": 0.24347826086956523
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.04645207425374216,
+      "wer": 0.1628888631895455
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.046959768900162055,
+      "wer": 0.16300211416490487
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.048830955902505996,
+      "wer": 0.1671315993932443
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.05059132720105125,
+      "wer": 0.17792985457656116
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.03795298726738491,
+      "wer": 0.1437037037037037
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.04497716894977169,
+      "wer": 0.15808617795187466
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.04040715607649599,
+      "wer": 0.13906678865507777
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.046296296296296294,
+      "wer": 0.17197452229299362
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.04364923318914668,
+      "wer": 0.1614420062695925
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.04329450572105968,
+      "wer": 0.14814814814814814
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.03945834951995116,
+      "wer": 0.1546942866688941
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.059343434343434344,
+      "wer": 0.17747949291573453
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.03408917972911277,
+      "wer": 0.12053571428571429
+    }
+  ]
+}

cv7_transcriptions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cea97d31b1ecb4c149b2fdf297b155571b4b9b7731db135f17756dcb46c5f5e
+size 125438911

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "processor_class": "Wav2Vec2Processor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95411e8343a965b718ba44b1710cc891b3c5bf213521ecbe61d58144e131ac68
+size 1262066801

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "do_lower_case": false, "word_delimiter_token": "|", "replace_word_delimiter_char": " ", "special_tokens_map_file": null, "name_or_path": "/tmp/tmpxk2ovcyi", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2Processor"}

vocab.json

@@ -0,0 +1 @@
+{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "'": 5, "-": 6, "a": 7, "b": 8, "c": 9, "d": 10, "e": 11, "f": 12, "g": 13, "h": 14, "i": 15, "j": 16, "k": 17, "l": 18, "m": 19, "n": 20, "o": 21, "p": 22, "q": 23, "r": 24, "s": 25, "t": 26, "u": 27, "v": 28, "w": 29, "x": 30, "y": 31, "z": 32, "á": 33, "é": 34, "í": 35, "ñ": 36, "ó": 37, "ö": 38, "ú": 39, "ü": 40}