jonatasgrosman/exp_w2v2r_en_vp-100k_age_teens-0_sixties-10_s539

.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-0_sixties-10_s539
+
+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": 0.17438602823485938,
+      "wer": 0.44968742185546384
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.17216377539828434,
+      "wer": 0.44529611523678025
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.1796260152364163,
+      "wer": 0.4620727472125754
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.12791687957758474,
+      "wer": 0.35764235764235763
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.14038785834738618,
+      "wer": 0.37320574162679426
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.14382083108472746,
+      "wer": 0.4039525691699605
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.2058767354734445,
+      "wer": 0.5076962765500043
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.17562724014336917,
+      "wer": 0.4807692307692308
+    },
+    {
+      "subset_type": "age",
+      "subset": "nineties",
+      "cer": 0.23076923076923078,
+      "wer": 0.5714285714285714
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.18561143784222267,
+      "wer": 0.4703115814226925
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.16046658259773014,
+      "wer": 0.43333333333333335
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.17227852323164944,
+      "wer": 0.44554403719708335
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.1872861109317492,
+      "wer": 0.4736891605925229
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.1651829871414441,
+      "wer": 0.4820359281437126
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.17687048098082364,
+      "wer": 0.45745654162854527
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.175443917303929,
+      "wer": 0.4511764005852252
+    },
+    {
+      "subset_type": "accent",
+      "subset": "england",
+      "cer": 0.16171418756441086,
+      "wer": 0.4404170804369414
+    },
+    {
+      "subset_type": "accent",
+      "subset": "australia",
+      "cer": 0.11687335588124766,
+      "wer": 0.3552036199095023
+    },
+    {
+      "subset_type": "accent",
+      "subset": "malaysia",
+      "cer": 0.19487179487179487,
+      "wer": 0.4583333333333333
+    },
+    {
+      "subset_type": "accent",
+      "subset": "indian",
+      "cer": 0.22854055166735282,
+      "wer": 0.539386650631389
+    },
+    {
+      "subset_type": "accent",
+      "subset": "wales",
+      "cer": 0.09378531073446328,
+      "wer": 0.3611111111111111
+    },
+    {
+      "subset_type": "accent",
+      "subset": "philippines",
+      "cer": 0.23200475907198095,
+      "wer": 0.5536332179930796
+    },
+    {
+      "subset_type": "accent",
+      "subset": "hongkong",
+      "cer": 0.16758747697974216,
+      "wer": 0.4673913043478261
+    },
+    {
+      "subset_type": "accent",
+      "subset": "bermuda",
+      "cer": 0.08641975308641975,
+      "wer": 0.25
+    },
+    {
+      "subset_type": "accent",
+      "subset": "ireland",
+      "cer": 0.20685111989459815,
+      "wer": 0.48031496062992124
+    },
+    {
+      "subset_type": "accent",
+      "subset": "newzealand",
+      "cer": 0.10344827586206896,
+      "wer": 0.35
+    },
+    {
+      "subset_type": "accent",
+      "subset": "other",
+      "cer": 0.17843866171003717,
+      "wer": 0.54
+    },
+    {
+      "subset_type": "accent",
+      "subset": "us",
+      "cer": 0.1446726448751326,
+      "wer": 0.40042850699142984
+    },
+    {
+      "subset_type": "accent",
+      "subset": "african",
+      "cer": 0.20302586627623231,
+      "wer": 0.5189504373177842
+    },
+    {
+      "subset_type": "accent",
+      "subset": "scotland",
+      "cer": 0.09074733096085409,
+      "wer": 0.29545454545454547
+    },
+    {
+      "subset_type": "accent",
+      "subset": "singapore",
+      "cer": 0.2425629290617849,
+      "wer": 0.4935064935064935
+    },
+    {
+      "subset_type": "accent",
+      "subset": "southatlandtic",
+      "cer": 0.21978021978021978,
+      "wer": 0.5757575757575758
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canada",
+      "cer": 0.12769857433808554,
+      "wer": 0.37290167865707435
+    }
+  ],
+  "test": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.22007183063337854,
+      "wer": 0.5134048257372654
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.21835861078338598,
+      "wer": 0.5098926677874046
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.21267532914551243,
+      "wer": 0.5121951219512195
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.17058468957203135,
+      "wer": 0.4597315436241611
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.2694693314955203,
+      "wer": 0.5860058309037901
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.2340119165839126,
+      "wer": 0.5437744117366867
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.24,
+      "wer": 0.72
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.20894260840165665,
+      "wer": 0.48872180451127817
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.2507001966509743,
+      "wer": 0.5663716814159292
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.22052927024859664,
+      "wer": 0.4713656387665198
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.21830545366710227,
+      "wer": 0.5098279257563891
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.23387327345683273,
+      "wer": 0.5396415126144856
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.2286902286902287,
+      "wer": 0.5019455252918288
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.21430061541671014,
+      "wer": 0.5153137079988106
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.22067946559001464,
+      "wer": 0.5142645344234451
+    },
+    {
+      "subset_type": "accent",
+      "subset": "england",
+      "cer": 0.18845735744638797,
+      "wer": 0.47484454494064443
+    },
+    {
+      "subset_type": "accent",
+      "subset": "malaysia",
+      "cer": 0.1612541993281075,
+      "wer": 0.4140127388535032
+    },
+    {
+      "subset_type": "accent",
+      "subset": "australia",
+      "cer": 0.16381156316916487,
+      "wer": 0.43312101910828027
+    },
+    {
+      "subset_type": "accent",
+      "subset": "indian",
+      "cer": 0.2595054945054945,
+      "wer": 0.5783851712221175
+    },
+    {
+      "subset_type": "accent",
+      "subset": "wales",
+      "cer": 0.18493150684931506,
+      "wer": 0.4583333333333333
+    },
+    {
+      "subset_type": "accent",
+      "subset": "philippines",
+      "cer": 0.19230769230769232,
+      "wer": 0.46464646464646464
+    },
+    {
+      "subset_type": "accent",
+      "subset": "hongkong",
+      "cer": 0.30491329479768786,
+      "wer": 0.675
+    },
+    {
+      "subset_type": "accent",
+      "subset": "bermuda",
+      "cer": 0.24044943820224718,
+      "wer": 0.5119047619047619
+    },
+    {
+      "subset_type": "accent",
+      "subset": "ireland",
+      "cer": 0.18432510885341075,
+      "wer": 0.4472573839662447
+    },
+    {
+      "subset_type": "accent",
+      "subset": "newzealand",
+      "cer": 0.14285714285714285,
+      "wer": 0.39705882352941174
+    },
+    {
+      "subset_type": "accent",
+      "subset": "other",
+      "cer": 0.29875195007800315,
+      "wer": 0.6138996138996139
+    },
+    {
+      "subset_type": "accent",
+      "subset": "us",
+      "cer": 0.19956187797490263,
+      "wer": 0.4831117224514749
+    },
+    {
+      "subset_type": "accent",
+      "subset": "african",
+      "cer": 0.2547642928786359,
+      "wer": 0.5375722543352601
+    },
+    {
+      "subset_type": "accent",
+      "subset": "scotland",
+      "cer": 0.22019635343618513,
+      "wer": 0.5602836879432624
+    },
+    {
+      "subset_type": "accent",
+      "subset": "singapore",
+      "cer": 0.3048128342245989,
+      "wer": 0.6363636363636364
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canada",
+      "cer": 0.20854021847070506,
+      "wer": 0.4550561797752809
+    }
+  ]
+}

cv7_transcriptions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5a78527904c847e4c32db8e3f91e2eaaefc2a92b987d2ae70cbf8373ac42d60
+size 120942314

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 sha256:7070d4cb4139fc6bcc968a95c46e0b70589ed47dd1575f557f804927e1e3a92c
+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/tmpkcs2izbc", "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}