Added model files

README.md

@@ -0,0 +1,70 @@
+---
+language:
+- es
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- common_voice
+- generated_from_trainer
+datasets:
+- common_voice
+model-index:
+- name: wav2vec2-cls-r-300m-es
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# wav2vec2-cls-r-300m-es
+
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the COMMON_VOICE - ES dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5160
+- Wer: 0.4016
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0003
+- train_batch_size: 16
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 8.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| 3.1277        | 1.14  | 500  | 2.0259          | 0.9999 |
+| 1.4111        | 2.28  | 1000 | 1.1251          | 0.8894 |
+| 0.8461        | 3.42  | 1500 | 0.8205          | 0.7244 |
+| 0.5042        | 4.57  | 2000 | 0.6116          | 0.5463 |
+| 0.3072        | 5.71  | 2500 | 0.5507          | 0.4506 |
+| 0.2181        | 6.85  | 3000 | 0.5213          | 0.4177 |
+| 0.1608        | 7.99  | 3500 | 0.5161          | 0.4019 |
+
+
+### Framework versions
+
+- Transformers 4.17.0.dev0
+- Pytorch 1.10.2+cu102
+- Datasets 1.18.2.dev0
+- Tokenizers 0.11.0

added_tokens.json

@@ -0,0 +1 @@
+{"<s>": 68, "</s>": 69}

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 8.0,
+    "eval_loss": 0.5159734487533569,
+    "eval_runtime": 175.5351,
+    "eval_samples": 3500,
+    "eval_samples_per_second": 19.939,
+    "eval_steps_per_second": 2.495,
+    "eval_wer": 0.4015755133948706,
+    "train_loss": 0.9386324367770865,
+    "train_runtime": 4216.4813,
+    "train_samples": 7000,
+    "train_samples_per_second": 13.281,
+    "train_steps_per_second": 0.831
+}

checkpoint-3450/config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "activation_dropout": 0.0,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.0,
+  "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": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.0,
+  "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": 67,
+  "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": "4.17.0.dev0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 70,
+  "xvector_output_dim": 512
+}

checkpoint-3450/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43266183f86e759c109db8d25ef69db7700d7c0c9a89a1d78ff11dda197374f5
+size 2490632977

checkpoint-3450/preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

checkpoint-3450/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fea34cae97ec757af794f7419a4e2f7bbd645ed8cc144687e02e3afc30c5b36a
+size 1262210673

checkpoint-3450/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2245792f4de9a99b157a2197b3c8813d18ae12216a3507290241bc22f3e987df
+size 14567

checkpoint-3450/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c7e0b4359f8962098d8fd507a364d1ce043548ef8362daf6cdbbc79fb55c984
+size 559

checkpoint-3450/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06bb2293e52a1e21a65c4c31cf2cd9297e007731fa2cc5decd28c3df11b31ec2
+size 623

checkpoint-3450/trainer_state.json

@@ -0,0 +1,106 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 7.876712328767123,
+  "global_step": 3450,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.0002574486301369863,
+      "loss": 3.1277,
+      "step": 500
+    },
+    {
+      "epoch": 1.14,
+      "eval_loss": 2.02591609954834,
+      "eval_runtime": 179.7935,
+      "eval_samples_per_second": 19.467,
+      "eval_steps_per_second": 2.436,
+      "eval_wer": 0.9999398659009592,
+      "step": 500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.0002146404109589041,
+      "loss": 1.4111,
+      "step": 1000
+    },
+    {
+      "epoch": 2.28,
+      "eval_loss": 1.125086784362793,
+      "eval_runtime": 179.8421,
+      "eval_samples_per_second": 19.462,
+      "eval_steps_per_second": 2.435,
+      "eval_wer": 0.889383324814336,
+      "step": 1000
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 0.00017183219178082193,
+      "loss": 0.8461,
+      "step": 1500
+    },
+    {
+      "epoch": 3.42,
+      "eval_loss": 0.8204647302627563,
+      "eval_runtime": 183.3614,
+      "eval_samples_per_second": 19.088,
+      "eval_steps_per_second": 2.389,
+      "eval_wer": 0.724375357046213,
+      "step": 1500
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 0.00012902397260273972,
+      "loss": 0.5042,
+      "step": 2000
+    },
+    {
+      "epoch": 4.57,
+      "eval_loss": 0.6116402745246887,
+      "eval_runtime": 180.1904,
+      "eval_samples_per_second": 19.424,
+      "eval_steps_per_second": 2.431,
+      "eval_wer": 0.5462882227367029,
+      "step": 2000
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 8.621575342465752e-05,
+      "loss": 0.3072,
+      "step": 2500
+    },
+    {
+      "epoch": 5.71,
+      "eval_loss": 0.5506712198257446,
+      "eval_runtime": 180.8043,
+      "eval_samples_per_second": 19.358,
+      "eval_steps_per_second": 2.423,
+      "eval_wer": 0.45055473706365196,
+      "step": 2500
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 4.340753424657534e-05,
+      "loss": 0.2181,
+      "step": 3000
+    },
+    {
+      "epoch": 6.85,
+      "eval_loss": 0.5213309526443481,
+      "eval_runtime": 179.0608,
+      "eval_samples_per_second": 19.546,
+      "eval_steps_per_second": 2.446,
+      "eval_wer": 0.4177215189873418,
+      "step": 3000
+    }
+  ],
+  "max_steps": 3504,
+  "num_train_epochs": 8,
+  "total_flos": 7.313890427943137e+18,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3450/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a24f7a88f43f1dc912b86bff41a4f4a9653d396ed16824971198ad28a36c4c8
+size 3055

config.json

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "./wav2vec2-cls-r-300m-es",
+  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
   "activation_dropout": 0.0,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
   "add_adapter": false,
   "apply_spec_augment": true,
   "architectures": [
-    "Wav2Vec2Model"
+    "Wav2Vec2ForCTC"
   ],
   "attention_dropout": 0.0,
   "bos_token_id": 1,

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_loss": 0.5159734487533569,
+    "eval_runtime": 175.5351,
+    "eval_samples": 3500,
+    "eval_samples_per_second": 19.939,
+    "eval_steps_per_second": 2.495,
+    "eval_wer": 0.4015755133948706
+}

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f568ca58788b1feb21fa78f4d182d7becf4ff4b3b4bfaf6eaf9878b6730aef1
-size 1261910005
+oid sha256:8aec1df821abafa72292bb7a6417d56e0e31d55cf5b4e974671f377d5d8241d1
+size 1262210673

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

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": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./wav2vec2-cls-r-300m-es", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.9386324367770865,
+    "train_runtime": 4216.4813,
+    "train_samples": 7000,
+    "train_samples_per_second": 13.281,
+    "train_steps_per_second": 0.831
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "global_step": 3504,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.0002574486301369863,
+      "loss": 3.1277,
+      "step": 500
+    },
+    {
+      "epoch": 1.14,
+      "eval_loss": 2.02591609954834,
+      "eval_runtime": 179.7935,
+      "eval_samples_per_second": 19.467,
+      "eval_steps_per_second": 2.436,
+      "eval_wer": 0.9999398659009592,
+      "step": 500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.0002146404109589041,
+      "loss": 1.4111,
+      "step": 1000
+    },
+    {
+      "epoch": 2.28,
+      "eval_loss": 1.125086784362793,
+      "eval_runtime": 179.8421,
+      "eval_samples_per_second": 19.462,
+      "eval_steps_per_second": 2.435,
+      "eval_wer": 0.889383324814336,
+      "step": 1000
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 0.00017183219178082193,
+      "loss": 0.8461,
+      "step": 1500
+    },
+    {
+      "epoch": 3.42,
+      "eval_loss": 0.8204647302627563,
+      "eval_runtime": 183.3614,
+      "eval_samples_per_second": 19.088,
+      "eval_steps_per_second": 2.389,
+      "eval_wer": 0.724375357046213,
+      "step": 1500
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 0.00012902397260273972,
+      "loss": 0.5042,
+      "step": 2000
+    },
+    {
+      "epoch": 4.57,
+      "eval_loss": 0.6116402745246887,
+      "eval_runtime": 180.1904,
+      "eval_samples_per_second": 19.424,
+      "eval_steps_per_second": 2.431,
+      "eval_wer": 0.5462882227367029,
+      "step": 2000
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 8.621575342465752e-05,
+      "loss": 0.3072,
+      "step": 2500
+    },
+    {
+      "epoch": 5.71,
+      "eval_loss": 0.5506712198257446,
+      "eval_runtime": 180.8043,
+      "eval_samples_per_second": 19.358,
+      "eval_steps_per_second": 2.423,
+      "eval_wer": 0.45055473706365196,
+      "step": 2500
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 4.340753424657534e-05,
+      "loss": 0.2181,
+      "step": 3000
+    },
+    {
+      "epoch": 6.85,
+      "eval_loss": 0.5213309526443481,
+      "eval_runtime": 179.0608,
+      "eval_samples_per_second": 19.546,
+      "eval_steps_per_second": 2.446,
+      "eval_wer": 0.4177215189873418,
+      "step": 3000
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 5.993150684931506e-07,
+      "loss": 0.1608,
+      "step": 3500
+    },
+    {
+      "epoch": 7.99,
+      "eval_loss": 0.5160985589027405,
+      "eval_runtime": 181.598,
+      "eval_samples_per_second": 19.273,
+      "eval_steps_per_second": 2.412,
+      "eval_wer": 0.40187618389007485,
+      "step": 3500
+    },
+    {
+      "epoch": 8.0,
+      "step": 3504,
+      "total_flos": 7.417092095436165e+18,
+      "train_loss": 0.9386324367770865,
+      "train_runtime": 4216.4813,
+      "train_samples_per_second": 13.281,
+      "train_steps_per_second": 0.831
+    }
+  ],
+  "max_steps": 3504,
+  "num_train_epochs": 8,
+  "total_flos": 7.417092095436165e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a24f7a88f43f1dc912b86bff41a4f4a9653d396ed16824971198ad28a36c4c8
+size 3055

vocab.json

@@ -0,0 +1 @@
+{"!": 1, "\"": 2, "'": 3, ",": 4, "-": 5, ".": 6, ":": 7, ";": 8, "?": 9, "_": 10, "`": 11, "a": 12, "b": 13, "c": 14, "d": 15, "e": 16, "f": 17, "g": 18, "h": 19, "i": 20, "j": 21, "k": 22, "l": 23, "m": 24, "n": 25, "o": 26, "p": 27, "q": 28, "r": 29, "s": 30, "t": 31, "u": 32, "v": 33, "w": 34, "x": 35, "y": 36, "z": 37, "¡": 38, "¿": 39, "ß": 40, "à": 41, "á": 42, "â": 43, "ã": 44, "ä": 45, "é": 46, "ê": 47, "í": 48, "ñ": 49, "ó": 50, "ö": 51, "ú": 52, "ü": 53, "ā": 54, "ł": 55, "ō": 56, "ź": 57, "‘": 58, "’": 59, "“": 60, "…": 61, "ひ": 62, "ヒ": 63, "比": 64, "": 65, "|": 0, "[UNK]": 66, "[PAD]": 67}