jonatasgrosman/exp_w2v2r_es_vp-100k_age_teens-0_sixties-10_s666

.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-0_sixties-10_s666
+
+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": 0.06954169581912438,
+      "wer": 0.23376992038447827
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.06946034805283147,
+      "wer": 0.23332115020102168
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.06805737214303133,
+      "wer": 0.23443901149729088
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.08385246712152675,
+      "wer": 0.23422897196261683
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.0532258064516129,
+      "wer": 0.19811320754716982
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.06614394853236831,
+      "wer": 0.22351524879614768
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.07306357527874179,
+      "wer": 0.24340219069322944
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.06930311068242102,
+      "wer": 0.24755606670500288
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.06146737608513021,
+      "wer": 0.2130115424973767
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.06933193800203644,
+      "wer": 0.23299719270272334
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.0703847474871848,
+      "wer": 0.23437560540896588
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.09390096618357488,
+      "wer": 0.3508137432188065
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.06808421249183083,
+      "wer": 0.2352614787071716
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.06918807810894141,
+      "wer": 0.23327184836534792
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.09393482546677491,
+      "wer": 0.2864976010966415
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.061975591151792526,
+      "wer": 0.20728291316526612
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.06382978723404255,
+      "wer": 0.21100917431192662
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.07096548098826003,
+      "wer": 0.2277951933124347
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.06943827708703375,
+      "wer": 0.23889074507183428
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.06623265692453309,
+      "wer": 0.2304077401520387
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.06732367608644027,
+      "wer": 0.22779369627507162
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.07486059124589413,
+      "wer": 0.24448487605185354
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.07518616382416526,
+      "wer": 0.2225866916588566
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.07985879209999046,
+      "wer": 0.2603926096997691
+    }
+  ],
+  "test": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.08095321051958419,
+      "wer": 0.2565754266346897
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.08125310739144846,
+      "wer": 0.2569211669770329
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.0813573652050267,
+      "wer": 0.25448275862068964
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.07287213369607462,
+      "wer": 0.23990772779700115
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.14760914760914762,
+      "wer": 0.4024390243902439
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.06552240839122735,
+      "wer": 0.21921679450948728
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.08162774381076476,
+      "wer": 0.2631389488840893
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.04697986577181208,
+      "wer": 0.18518518518518517
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.09220192761339893,
+      "wer": 0.30565730565730564
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.07234115284365253,
+      "wer": 0.22728442728442727
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.0813772387755867,
+      "wer": 0.2572402308198753
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.07694017374834514,
+      "wer": 0.25118538221348446
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.14347202295552366,
+      "wer": 0.3565217391304348
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.0809554005495667,
+      "wer": 0.253276955602537
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.08175160312996156,
+      "wer": 0.2583521402937586
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.06307834669956817,
+      "wer": 0.20768526989935956
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.06483867642511082,
+      "wer": 0.20886460230722526
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.060699588477366256,
+      "wer": 0.2229299363057325
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.061704211557296766,
+      "wer": 0.2325925925925926
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.09088918090232151,
+      "wer": 0.28186484174508125
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.07776255707762557,
+      "wer": 0.23922775601566873
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.09583333333333334,
+      "wer": 0.28560775540641314
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.06726233420278595,
+      "wer": 0.2368860674908119
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.053873078679044284,
+      "wer": 0.1982142857142857
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.07628784899724735,
+      "wer": 0.25783699059561127
+    }
+  ]
+}

cv7_transcriptions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c9c0f4a9d73b9f44c2859a231451bda2bbb19c410c8d9636bd997233aed855c
+size 124765764

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:f500bf5bce35ec28a468b8ad7789430e0b17a68d0b77a2771e7c7cb5fb2420ec
+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/tmphutng9hz", "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}