jonatasgrosman/exp_w2v2r_es_vp-100k_age_teens-10_sixties-0_s261

.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_vp-100k_age_teens-10_sixties-0_s261
+
+Fine-tuned [facebook/wav2vec2-large-100k-voxpopuli](https://huggingface.co/facebook/wav2vec2-large-100k-voxpopuli) 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 +1,116 @@
+{
+  "_name_or_path": "facebook/wav2vec2-large-100k-voxpopuli",
+  "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_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "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_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
+  "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": 0.05236093589030229,
+      "wer": 0.18470669008040771
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.052210759027266025,
+      "wer": 0.1846424533995189
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.05265024717905518,
+      "wer": 0.1871283203383111
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.07623850489469,
+      "wer": 0.20619158878504673
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.054838709677419356,
+      "wer": 0.2358490566037736
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.05126658624849216,
+      "wer": 0.18138041733547353
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.05329936744689753,
+      "wer": 0.18628243769756672
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.051603568844948156,
+      "wer": 0.18545140885566416
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.045505460655278634,
+      "wer": 0.16915005246589718
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.052156635365991626,
+      "wer": 0.1845267880617775
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.05371689272822713,
+      "wer": 0.18644658839939557
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.0625,
+      "wer": 0.2423146473779385
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.05006535337503501,
+      "wer": 0.17700097100846165
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.052070417398799855,
+      "wer": 0.18464571937412544
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.06053577641192161,
+      "wer": 0.21110349554489377
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.04776887871853547,
+      "wer": 0.17198879551820728
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.043286867204695524,
+      "wer": 0.1743119266055046
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.06623444892237602,
+      "wer": 0.2309299895506792
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.05073268206039076,
+      "wer": 0.18242565987303708
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.050372832836992405,
+      "wer": 0.17733241188666207
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.05271906910472572,
+      "wer": 0.18624641833810887
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.05198227790084791,
+      "wer": 0.1773936775073914
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.07174313395788294,
+      "wer": 0.20618556701030927
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.053048373246827596,
+      "wer": 0.18013856812933027
+    }
+  ],
+  "test": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.059869497527994746,
+      "wer": 0.20177855297669778
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.06036237363274776,
+      "wer": 0.20263035381750466
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.05653126321048384,
+      "wer": 0.2006896551724138
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.05557714729887291,
+      "wer": 0.1914648212226067
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.11850311850311851,
+      "wer": 0.3048780487804878
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.054079825636834215,
+      "wer": 0.17844166330238193
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.05835421469379573,
+      "wer": 0.1994240460763139
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.053691275167785234,
+      "wer": 0.14814814814814814
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.06524229965626474,
+      "wer": 0.22995522995522996
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.050549074904926716,
+      "wer": 0.17734877734877735
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.06040897358513002,
+      "wer": 0.20283490182409666
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.05599953615569707,
+      "wer": 0.1931883890366601
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.09684361549497848,
+      "wer": 0.2826086956521739
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.05745790178256887,
+      "wer": 0.20042283298097252
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.06056337332618911,
+      "wer": 0.2034703466646935
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.04549660703269587,
+      "wer": 0.16285452881976212
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.05185032470879291,
+      "wer": 0.17304189435336975
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.06378600823045268,
+      "wer": 0.2229299363057325
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.05386875612144956,
+      "wer": 0.2162962962962963
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.06672506935319025,
+      "wer": 0.21685201026518391
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.05328767123287671,
+      "wer": 0.18270844991606044
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.07108585858585859,
+      "wer": 0.21774794929157346
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.04794938675842167,
+      "wer": 0.17741396592048111
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.04139400395677979,
+      "wer": 0.16517857142857142
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.05033425088478175,
+      "wer": 0.1810344827586207
+    }
+  ]
+}

cv7_transcriptions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4302ee40b9463f56d391a6976af8ee938b47d76d80709848e78aae573649eb8d
+size 125319452

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 sha256:7cdcd3176b083ec6d25525a2f54ade91ffe2cd1275a79c9ee262eae30ca78a28
+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/tmp_qvebhu9", "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}