jonatasgrosman/exp_w2v2r_en_vp-100k_age_teens-2_sixties-8_s769

.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:
+- en
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- en
+datasets:
+- mozilla-foundation/common_voice_7_0
+---
+# exp_w2v2r_en_vp-100k_age_teens-2_sixties-8_s769
+
+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 (en)](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": 33,
+  "xvector_output_dim": 512
+}

cv7_evaluation.json

@@ -0,0 +1,390 @@
+{
+  "validation": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.15981849939736129,
+      "wer": 0.423624656164041
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.15773843503337262,
+      "wer": 0.41945740194797926
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.16697097525656363,
+      "wer": 0.4403216962164138
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.12110373019928462,
+      "wer": 0.35264735264735264
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.1391231028667791,
+      "wer": 0.38516746411483255
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.13464651915812195,
+      "wer": 0.38181818181818183
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.1865422757658121,
+      "wer": 0.4716656634276378
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.18996415770609318,
+      "wer": 0.4807692307692308
+    },
+    {
+      "subset_type": "age",
+      "subset": "nineties",
+      "cer": 0.19230769230769232,
+      "wer": 0.5714285714285714
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.16715676333400933,
+      "wer": 0.44238683127572015
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.1525220680958386,
+      "wer": 0.4164835164835165
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.15784308694871335,
+      "wer": 0.41968370537879973
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.1715793891651062,
+      "wer": 0.44702562896778747
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.1493570722057369,
+      "wer": 0.437125748502994
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.16390286073561772,
+      "wer": 0.429551692589204
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.16052511860661597,
+      "wer": 0.42441424981274195
+    },
+    {
+      "subset_type": "accent",
+      "subset": "england",
+      "cer": 0.15063552043971143,
+      "wer": 0.423038728897716
+    },
+    {
+      "subset_type": "accent",
+      "subset": "australia",
+      "cer": 0.11161217587373168,
+      "wer": 0.3235294117647059
+    },
+    {
+      "subset_type": "accent",
+      "subset": "malaysia",
+      "cer": 0.18717948717948718,
+      "wer": 0.4166666666666667
+    },
+    {
+      "subset_type": "accent",
+      "subset": "indian",
+      "cer": 0.20512556607657473,
+      "wer": 0.5033072760072159
+    },
+    {
+      "subset_type": "accent",
+      "subset": "wales",
+      "cer": 0.09717514124293786,
+      "wer": 0.3194444444444444
+    },
+    {
+      "subset_type": "accent",
+      "subset": "philippines",
+      "cer": 0.2070196311719215,
+      "wer": 0.5432525951557093
+    },
+    {
+      "subset_type": "accent",
+      "subset": "hongkong",
+      "cer": 0.19337016574585636,
+      "wer": 0.4673913043478261
+    },
+    {
+      "subset_type": "accent",
+      "subset": "bermuda",
+      "cer": 0.06172839506172839,
+      "wer": 0.125
+    },
+    {
+      "subset_type": "accent",
+      "subset": "ireland",
+      "cer": 0.18840579710144928,
+      "wer": 0.49606299212598426
+    },
+    {
+      "subset_type": "accent",
+      "subset": "newzealand",
+      "cer": 0.059113300492610835,
+      "wer": 0.275
+    },
+    {
+      "subset_type": "accent",
+      "subset": "other",
+      "cer": 0.1691449814126394,
+      "wer": 0.48
+    },
+    {
+      "subset_type": "accent",
+      "subset": "us",
+      "cer": 0.13587889306720663,
+      "wer": 0.38373928732521423
+    },
+    {
+      "subset_type": "accent",
+      "subset": "african",
+      "cer": 0.19619326500732065,
+      "wer": 0.4839650145772595
+    },
+    {
+      "subset_type": "accent",
+      "subset": "scotland",
+      "cer": 0.09252669039145907,
+      "wer": 0.2840909090909091
+    },
+    {
+      "subset_type": "accent",
+      "subset": "singapore",
+      "cer": 0.21624713958810068,
+      "wer": 0.4025974025974026
+    },
+    {
+      "subset_type": "accent",
+      "subset": "southatlandtic",
+      "cer": 0.17582417582417584,
+      "wer": 0.5151515151515151
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canada",
+      "cer": 0.12871690427698573,
+      "wer": 0.39568345323741005
+    }
+  ],
+  "test": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.2027486568137235,
+      "wer": 0.48652978486889426
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.20130906010244937,
+      "wer": 0.4837404249168955
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.1961104896308407,
+      "wer": 0.49073170731707316
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.15913200723327306,
+      "wer": 0.41946308724832215
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.25,
+      "wer": 0.5335276967930029
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.21415094339622642,
+      "wer": 0.5137658378584358
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.12571428571428572,
+      "wer": 0.48
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.19296762741949117,
+      "wer": 0.45206766917293234
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.22871104225016387,
+      "wer": 0.5299523485364194
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.20970328789093826,
+      "wer": 0.44493392070484583
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.20124383071634933,
+      "wer": 0.48363279675773496
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.21439204976782247,
+      "wer": 0.5080069674664269
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.2203742203742204,
+      "wer": 0.4980544747081712
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.1976635026598519,
+      "wer": 0.4852809991079393
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.20330357406236493,
+      "wer": 0.48775063693809106
+    },
+    {
+      "subset_type": "accent",
+      "subset": "england",
+      "cer": 0.1770925980828145,
+      "wer": 0.44714527981910684
+    },
+    {
+      "subset_type": "accent",
+      "subset": "malaysia",
+      "cer": 0.15789473684210525,
+      "wer": 0.39490445859872614
+    },
+    {
+      "subset_type": "accent",
+      "subset": "australia",
+      "cer": 0.15899357601713063,
+      "wer": 0.4171974522292994
+    },
+    {
+      "subset_type": "accent",
+      "subset": "indian",
+      "cer": 0.22796703296703297,
+      "wer": 0.5315739868049011
+    },
+    {
+      "subset_type": "accent",
+      "subset": "wales",
+      "cer": 0.1917808219178082,
+      "wer": 0.4583333333333333
+    },
+    {
+      "subset_type": "accent",
+      "subset": "philippines",
+      "cer": 0.1956521739130435,
+      "wer": 0.47474747474747475
+    },
+    {
+      "subset_type": "accent",
+      "subset": "hongkong",
+      "cer": 0.29190751445086704,
+      "wer": 0.6083333333333333
+    },
+    {
+      "subset_type": "accent",
+      "subset": "bermuda",
+      "cer": 0.2202247191011236,
+      "wer": 0.5476190476190477
+    },
+    {
+      "subset_type": "accent",
+      "subset": "ireland",
+      "cer": 0.1734397677793904,
+      "wer": 0.4092827004219409
+    },
+    {
+      "subset_type": "accent",
+      "subset": "newzealand",
+      "cer": 0.14150943396226415,
+      "wer": 0.375
+    },
+    {
+      "subset_type": "accent",
+      "subset": "other",
+      "cer": 0.27301092043681746,
+      "wer": 0.583011583011583
+    },
+    {
+      "subset_type": "accent",
+      "subset": "us",
+      "cer": 0.18525530073561228,
+      "wer": 0.45529573590096284
+    },
+    {
+      "subset_type": "accent",
+      "subset": "african",
+      "cer": 0.24974924774322968,
+      "wer": 0.5202312138728323
+    },
+    {
+      "subset_type": "accent",
+      "subset": "scotland",
+      "cer": 0.2061711079943899,
+      "wer": 0.48226950354609927
+    },
+    {
+      "subset_type": "accent",
+      "subset": "singapore",
+      "cer": 0.2727272727272727,
+      "wer": 0.6060606060606061
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canada",
+      "cer": 0.20431976166832175,
+      "wer": 0.45646067415730335
+    }
+  ]
+}

cv7_transcriptions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fd368d4c4e9670812d222de81af160122bc9052474f87cf7fb3957a642c1a3f
+size 121544986

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:acd77c71f790fbe4bb8ad02c221761761841b40cf85fc729baed9488f5cfc473
+size 1262034033

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/tmplle98a1i", "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}