Upload noLM model

README.md

@@ -0,0 +1,84 @@
+---
+license: apache-2.0
+tags:
+- generated_from_trainer
+model-index:
+- name: wav2vec-base-Millad_TIMIT
+  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. -->
+
+# wav2vec-base-Millad_TIMIT
+
+This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.3772
+- Wer: 0.6859
+- Cer: 0.3217
+
+## 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: 0.0001
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 5000
+- num_epochs: 60
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Wer    | Cer    |
+|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
+| No log        | 2.36  | 2000  | 2.6233          | 1.0130 | 0.6241 |
+| No log        | 4.73  | 4000  | 2.2206          | 0.9535 | 0.5032 |
+| No log        | 7.09  | 6000  | 2.3036          | 0.9368 | 0.5063 |
+| 1.235         | 9.46  | 8000  | 1.9932          | 0.9275 | 0.5032 |
+| 1.235         | 11.82 | 10000 | 2.0207          | 0.8922 | 0.4498 |
+| 1.235         | 14.18 | 12000 | 1.6171          | 0.7993 | 0.3976 |
+| 1.235         | 16.55 | 14000 | 1.6729          | 0.8309 | 0.4209 |
+| 0.2779        | 18.91 | 16000 | 1.7043          | 0.8141 | 0.4340 |
+| 0.2779        | 21.28 | 18000 | 1.7426          | 0.7658 | 0.3960 |
+| 0.2779        | 23.64 | 20000 | 1.5230          | 0.7361 | 0.3830 |
+| 0.2779        | 26.0  | 22000 | 1.4286          | 0.7658 | 0.3794 |
+| 0.1929        | 28.37 | 24000 | 1.4450          | 0.7379 | 0.3644 |
+| 0.1929        | 30.73 | 26000 | 1.5922          | 0.7491 | 0.3826 |
+| 0.1929        | 33.1  | 28000 | 1.4443          | 0.7454 | 0.3617 |
+| 0.1929        | 35.46 | 30000 | 1.5450          | 0.7268 | 0.3621 |
+| 0.1394        | 37.83 | 32000 | 1.9268          | 0.7491 | 0.3763 |
+| 0.1394        | 40.19 | 34000 | 1.7094          | 0.7342 | 0.3783 |
+| 0.1394        | 42.55 | 36000 | 1.4024          | 0.7082 | 0.3494 |
+| 0.1394        | 44.92 | 38000 | 1.4467          | 0.6840 | 0.3395 |
+| 0.104         | 47.28 | 40000 | 1.4145          | 0.6933 | 0.3407 |
+| 0.104         | 49.65 | 42000 | 1.3901          | 0.6970 | 0.3403 |
+| 0.104         | 52.01 | 44000 | 1.3589          | 0.6636 | 0.3348 |
+| 0.104         | 54.37 | 46000 | 1.3716          | 0.6952 | 0.3340 |
+| 0.0781        | 56.74 | 48000 | 1.4025          | 0.6896 | 0.3312 |
+| 0.0781        | 59.1  | 50000 | 1.3772          | 0.6859 | 0.3217 |
+
+
+### Framework versions
+
+- Transformers 4.17.0
+- Pytorch 1.12.0+cu113
+- Datasets 1.18.3
+- Tokenizers 0.12.1

config.json

@@ -0,0 +1,117 @@
+{
+  "_name_or_path": "facebook/wav2vec2-base",
+  "activation_dropout": 0.0,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": false,
+  "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": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.0,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "freeze_feat_extract_train": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "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",
+  "no_mask_channel_overlap": false,
+  "no_mask_time_overlap": false,
+  "num_adapter_layers": 3,
+  "num_attention_heads": 12,
+  "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": 12,
+  "num_negatives": 100,
+  "output_hidden_size": 768,
+  "pad_token_id": 29,
+  "proj_codevector_dim": 256,
+  "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",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32,
+  "xvector_output_dim": 512
+}

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7660bb50b2610d533692541a858ad60a53f6ec386962b573063767d6c57b143
+size 377671575

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": " ", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fb6d7781d95ef03add3c711bad998ac868e58302c0baa3c2f1bbb146225d257
+size 3055

vocab.json

@@ -0,0 +1 @@
+{"y": 0, "r": 1, "m": 2, "j": 3, "a": 4, "g": 5, "q": 6, "s": 7, "n": 8, "l": 9, "d": 10, "i": 11, "e": 12, "o": 13, "u": 14, "z": 15, "t": 16, "x": 17, "f": 18, "p": 19, "'": 21, "k": 22, "c": 23, "b": 24, "h": 25, "v": 26, "w": 27, "[UNK]": 28, "[PAD]": 29, "|": 20}