jonatasgrosman/exp_w2v2r_es_xls-r_age_teens-5_sixties-5_s950

.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-5_sixties-5_s950
+
+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": 0.03740255733622746,
+      "wer": 0.13844914773102365
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.03749787427016609,
+      "wer": 0.13839471648717797
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.03859312313864481,
+      "wer": 0.14460045578181283
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.029823578829459536,
+      "wer": 0.11458551941238194
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.036711077612006475,
+      "wer": 0.1398176291793313
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.03524996649242729,
+      "wer": 0.12560192616372393
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.04687036487689113,
+      "wer": 0.15537383177570094
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.029838709677419355,
+      "wer": 0.14622641509433962
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.03882324571979744,
+      "wer": 0.14634847613571017
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.03744767507636611,
+      "wer": 0.13824454866971428
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.042270531400966184,
+      "wer": 0.1717902350813743
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.03792390592901257,
+      "wer": 0.1408423418187454
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.034403883857716366,
+      "wer": 0.13066999583853517
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.03739018002188111,
+      "wer": 0.13825372090064877
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.03607904085257549,
+      "wer": 0.1376545272302038
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.03677539995844588,
+      "wer": 0.13614374568071874
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.04415629928158402,
+      "wer": 0.1588296760710554
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.044648034326800416,
+      "wer": 0.1720356408498972
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.03008070432868672,
+      "wer": 0.11926605504587157
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.03873676175937411,
+      "wer": 0.1351039260969977
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.034229595728451565,
+      "wer": 0.12212885154061624
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.03349629516461691,
+      "wer": 0.128951557880373
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.03585846592258371,
+      "wer": 0.1421919770773639
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.046761149811834414,
+      "wer": 0.15979381443298968
+    }
+  ],
+  "test": [
+    {
+      "subset_type": "all",
+      "subset": "all",
+      "cer": 0.04285323654069216,
+      "wer": 0.15115057578340238
+    },
+    {
+      "subset_type": "age",
+      "subset": "",
+      "cer": 0.043093045657938346,
+      "wer": 0.15145872129112353
+    },
+    {
+      "subset_type": "age",
+      "subset": "twenties",
+      "cer": 0.04217700711636765,
+      "wer": 0.15490760739140869
+    },
+    {
+      "subset_type": "age",
+      "subset": "fourties",
+      "cer": 0.03760201683544845,
+      "wer": 0.13487773487773488
+    },
+    {
+      "subset_type": "age",
+      "subset": "thirties",
+      "cer": 0.04089005034395296,
+      "wer": 0.1496551724137931
+    },
+    {
+      "subset_type": "age",
+      "subset": "fifties",
+      "cer": 0.03855060618444354,
+      "wer": 0.13403310456197012
+    },
+    {
+      "subset_type": "age",
+      "subset": "sixties",
+      "cer": 0.038670812281383596,
+      "wer": 0.13610149942329874
+    },
+    {
+      "subset_type": "age",
+      "subset": "seventies",
+      "cer": 0.10602910602910603,
+      "wer": 0.2682926829268293
+    },
+    {
+      "subset_type": "age",
+      "subset": "teens",
+      "cer": 0.04832513311316304,
+      "wer": 0.17012617012617012
+    },
+    {
+      "subset_type": "age",
+      "subset": "eighties",
+      "cer": 0.013422818791946308,
+      "wer": 0.07407407407407407
+    },
+    {
+      "subset_type": "gender",
+      "subset": "",
+      "cer": 0.043132095029940816,
+      "wer": 0.15163626505557493
+    },
+    {
+      "subset_type": "gender",
+      "subset": "other",
+      "cer": 0.06456241032998565,
+      "wer": 0.20869565217391303
+    },
+    {
+      "subset_type": "gender",
+      "subset": "male",
+      "cer": 0.04086661577263898,
+      "wer": 0.1478547473112062
+    },
+    {
+      "subset_type": "gender",
+      "subset": "female",
+      "cer": 0.04142887338829,
+      "wer": 0.1471458773784355
+    },
+    {
+      "subset_type": "accent",
+      "subset": "",
+      "cer": 0.043293768472723954,
+      "wer": 0.15230308187502312
+    },
+    {
+      "subset_type": "accent",
+      "subset": "caribe",
+      "cer": 0.04533508541392904,
+      "wer": 0.15355004277159967
+    },
+    {
+      "subset_type": "accent",
+      "subset": "surpeninsular",
+      "cer": 0.034280117531831536,
+      "wer": 0.13333333333333333
+    },
+    {
+      "subset_type": "accent",
+      "subset": "mexicano",
+      "cer": 0.03922374429223744,
+      "wer": 0.1421376608841634
+    },
+    {
+      "subset_type": "accent",
+      "subset": "chileno",
+      "cer": 0.034392350400987044,
+      "wer": 0.12076852698993595
+    },
+    {
+      "subset_type": "accent",
+      "subset": "canario",
+      "cer": 0.04218106995884774,
+      "wer": 0.1592356687898089
+    },
+    {
+      "subset_type": "accent",
+      "subset": "americacentral",
+      "cer": 0.03919255472538996,
+      "wer": 0.14341692789968652
+    },
+    {
+      "subset_type": "accent",
+      "subset": "nortepeninsular",
+      "cer": 0.032573961447273474,
+      "wer": 0.11657559198542805
+    },
+    {
+      "subset_type": "accent",
+      "subset": "andino",
+      "cer": 0.036849991675453685,
+      "wer": 0.14734380220514534
+    },
+    {
+      "subset_type": "accent",
+      "subset": "rioplatense",
+      "cer": 0.054292929292929296,
+      "wer": 0.168530947054437
+    },
+    {
+      "subset_type": "accent",
+      "subset": "centrosurpeninsular",
+      "cer": 0.031349870643737635,
+      "wer": 0.12410714285714286
+    }
+  ]
+}

cv7_transcriptions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a64a676b8fd711e55697c45c2999cac8e57f182214c81e2ea9ec98fe37e2dee2
+size 125520390

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 sha256:61c19c7eb96405f61b2c6d6cc4e598ffcae24c393dc588187b08d2c4dd0b97a3
+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/tmpmmknpdb2", "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}