Model save

README.md

@@ -0,0 +1,114 @@
+---
+license: apache-2.0
+base_model: facebook/wav2vec2-large-960h
+tags:
+- generated_from_trainer
+metrics:
+- wer
+model-index:
+- name: Helldivers2ASR_V5
+  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. -->
+
+# Helldivers2ASR_V5
+
+This model is a fine-tuned version of [facebook/wav2vec2-large-960h](https://huggingface.co/facebook/wav2vec2-large-960h) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 297.9317
+- Wer: 0.3294
+- Cer: 0.1602
+
+## 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: 7e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.05
+- num_epochs: 50
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Wer    | Cer    |
+|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
+| 462.6882      | 1.0   | 510   | 344.5970        | 0.4610 | 0.2309 |
+| 362.2724      | 2.0   | 1020  | 296.6635        | 0.4070 | 0.2075 |
+| 315.8042      | 3.0   | 1530  | 275.4880        | 0.3980 | 0.1941 |
+| 282.2233      | 4.0   | 2040  | 244.6664        | 0.3790 | 0.1821 |
+| 258.098       | 5.0   | 2550  | 237.4571        | 0.3485 | 0.1716 |
+| 239.4842      | 6.0   | 3060  | 233.8290        | 0.3495 | 0.1669 |
+| 224.3915      | 7.0   | 3570  | 250.8993        | 0.3398 | 0.1648 |
+| 209.738       | 8.0   | 4080  | 230.4935        | 0.3306 | 0.1619 |
+| 198.5307      | 9.0   | 4590  | 240.8248        | 0.3237 | 0.1584 |
+| 185.1102      | 10.0  | 5100  | 236.7069        | 0.3212 | 0.1565 |
+| 175.6033      | 11.0  | 5610  | 235.4940        | 0.3229 | 0.1576 |
+| 165.3067      | 12.0  | 6120  | 244.4790        | 0.3257 | 0.1572 |
+| 155.9264      | 13.0  | 6630  | 240.0126        | 0.3272 | 0.1572 |
+| 149.6907      | 14.0  | 7140  | 255.7472        | 0.3218 | 0.1572 |
+| 140.2203      | 15.0  | 7650  | 254.3143        | 0.3297 | 0.1575 |
+| 135.4905      | 16.0  | 8160  | 251.0496        | 0.3249 | 0.1555 |
+| 127.9138      | 17.0  | 8670  | 251.7713        | 0.3166 | 0.1537 |
+| 122.4403      | 18.0  | 9180  | 249.0815        | 0.3217 | 0.1562 |
+| 116.9182      | 19.0  | 9690  | 255.4340        | 0.3245 | 0.1549 |
+| 112.3006      | 20.0  | 10200 | 258.1701        | 0.3288 | 0.1577 |
+| 107.9413      | 21.0  | 10710 | 266.8587        | 0.3271 | 0.1592 |
+| 104.409       | 22.0  | 11220 | 270.4631        | 0.3315 | 0.1580 |
+| 100.6648      | 23.0  | 11730 | 277.8313        | 0.3373 | 0.1602 |
+| 96.5402       | 24.0  | 12240 | 264.8732        | 0.3290 | 0.1567 |
+| 92.6933       | 25.0  | 12750 | 277.2248        | 0.3331 | 0.1603 |
+| 89.6166       | 26.0  | 13260 | 276.5128        | 0.3298 | 0.1583 |
+| 88.5449       | 27.0  | 13770 | 275.0455        | 0.3276 | 0.1557 |
+| 86.5652       | 28.0  | 14280 | 282.4394        | 0.3239 | 0.1569 |
+| 84.407        | 29.0  | 14790 | 284.3981        | 0.3263 | 0.1583 |
+| 81.4084       | 30.0  | 15300 | 285.0651        | 0.3304 | 0.1594 |
+| 78.9513       | 31.0  | 15810 | 289.1231        | 0.3274 | 0.1583 |
+| 77.8021       | 32.0  | 16320 | 282.9202        | 0.3327 | 0.1619 |
+| 76.3021       | 33.0  | 16830 | 290.6485        | 0.3278 | 0.1608 |
+| 73.3693       | 34.0  | 17340 | 291.9531        | 0.3329 | 0.1611 |
+| 73.2556       | 35.0  | 17850 | 288.5987        | 0.3267 | 0.1606 |
+| 70.036        | 36.0  | 18360 | 283.9933        | 0.3256 | 0.1588 |
+| 70.3601       | 37.0  | 18870 | 298.8391        | 0.3289 | 0.1598 |
+| 70.4905       | 38.0  | 19380 | 299.1319        | 0.3297 | 0.1587 |
+| 69.1989       | 39.0  | 19890 | 295.0404        | 0.3329 | 0.1607 |
+| 68.393        | 40.0  | 20400 | 293.8006        | 0.3320 | 0.1601 |
+| 65.682        | 41.0  | 20910 | 296.3048        | 0.3311 | 0.1592 |
+| 66.4088       | 42.0  | 21420 | 293.5957        | 0.3309 | 0.1606 |
+| 65.2298       | 43.0  | 21930 | 299.2396        | 0.3302 | 0.1600 |
+| 64.245        | 44.0  | 22440 | 295.2800        | 0.3291 | 0.1593 |
+| 65.9184       | 45.0  | 22950 | 297.8622        | 0.3298 | 0.1603 |
+| 64.5982       | 46.0  | 23460 | 297.0223        | 0.3291 | 0.1597 |
+| 64.105        | 47.0  | 23970 | 297.7955        | 0.3289 | 0.1597 |
+| 63.6948       | 48.0  | 24480 | 298.1255        | 0.3289 | 0.1599 |
+| 64.6734       | 49.0  | 24990 | 297.8814        | 0.3292 | 0.1601 |
+| 64.8948       | 50.0  | 25500 | 297.9317        | 0.3294 | 0.1602 |
+
+
+### Framework versions
+
+- Transformers 4.44.0
+- Pytorch 2.5.1+cu121
+- Datasets 3.6.0
+- Tokenizers 0.19.1

config.json

@@ -0,0 +1,109 @@
+{
+  "_name_or_path": "facebook/wav2vec2-large-960h",
+  "activation_dropout": 0.1,
+  "adapter_attn_dim": null,
+  "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": "sum",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.05,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.05,
+  "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": 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": "4.44.0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 32,
+  "xvector_output_dim": 512
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:656181f4c6482fafb3c79ed5db9c46ab690d2fc0f3e5253b1b005d646c28dd96
+size 1261897368

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "Wav2Vec2Processor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,50 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "do_lower_case": false,
+  "do_normalize": true,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "processor_class": "Wav2Vec2Processor",
+  "replace_word_delimiter_char": " ",
+  "return_attention_mask": false,
+  "target_lang": null,
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "unk_token": "<unk>",
+  "word_delimiter_token": "|"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d989275d83a6c80b5db22a282020a11fc19148b17eaefe1f51e96fea4ecad70c
+size 5240

vocab.json

@@ -0,0 +1,34 @@
+{
+  "'": 27,
+  "</s>": 2,
+  "<pad>": 0,
+  "<s>": 1,
+  "<unk>": 3,
+  "A": 7,
+  "B": 24,
+  "C": 19,
+  "D": 14,
+  "E": 5,
+  "F": 20,
+  "G": 21,
+  "H": 11,
+  "I": 10,
+  "J": 29,
+  "K": 26,
+  "L": 15,
+  "M": 17,
+  "N": 9,
+  "O": 8,
+  "P": 23,
+  "Q": 30,
+  "R": 13,
+  "S": 12,
+  "T": 6,
+  "U": 16,
+  "V": 25,
+  "W": 18,
+  "X": 28,
+  "Y": 22,
+  "Z": 31,
+  "|": 4
+}