updating the repo with the fine-tuned model

README.md

@@ -0,0 +1,82 @@
+---
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- experiments/data/uwb_atcc/train
+- generated_from_trainer
+metrics:
+- wer
+model-index:
+- name: 0.0ld_0.0ad_0.0attd_0.05fpd_0.075mtp_12mtl_0.0mfp_12mfl_1acc
+  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. -->
+
+# 0.0ld_0.0ad_0.0attd_0.05fpd_0.075mtp_12mtl_0.0mfp_12mfl_1acc
+
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the EXPERIMENTS/DATA/UWB_ATCC/TRAIN - NA dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8470
+- Wer: 0.1898
+
+## 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: 24
+- eval_batch_size: 12
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 1000
+- training_steps: 10000
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Wer    |
+|:-------------:|:-----:|:-----:|:---------------:|:------:|
+| No log        | 1.06  | 500   | 3.1697          | 1.0    |
+| 3.1489        | 2.12  | 1000  | 1.4184          | 0.5678 |
+| 3.1489        | 3.18  | 1500  | 0.8498          | 0.3366 |
+| 0.8499        | 4.25  | 2000  | 0.8089          | 0.2755 |
+| 0.8499        | 5.31  | 2500  | 0.7339          | 0.2963 |
+| 0.5901        | 6.37  | 3000  | 0.6376          | 0.2402 |
+| 0.5901        | 7.43  | 3500  | 0.6890          | 0.2336 |
+| 0.4724        | 8.49  | 4000  | 0.6844          | 0.2240 |
+| 0.4724        | 9.55  | 4500  | 0.6900          | 0.2222 |
+| 0.3981        | 10.62 | 5000  | 0.7051          | 0.2123 |
+| 0.3981        | 11.68 | 5500  | 0.6671          | 0.2095 |
+| 0.3436        | 12.74 | 6000  | 0.7425          | 0.2049 |
+| 0.3436        | 13.8  | 6500  | 0.7135          | 0.1994 |
+| 0.2925        | 14.86 | 7000  | 0.7350          | 0.2012 |
+| 0.2925        | 15.92 | 7500  | 0.7855          | 0.1945 |
+| 0.2525        | 16.99 | 8000  | 0.7933          | 0.1946 |
+| 0.2525        | 18.05 | 8500  | 0.8016          | 0.1915 |
+| 0.2285        | 19.11 | 9000  | 0.8284          | 0.1907 |
+| 0.2285        | 20.17 | 9500  | 0.8275          | 0.1902 |
+| 0.2025        | 21.23 | 10000 | 0.8470          | 0.1898 |
+
+
+### Framework versions
+
+- Transformers 4.24.0
+- Pytorch 1.13.0+cu117
+- Datasets 2.6.1
+- Tokenizers 0.13.2

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "</s>": 30,
+  "<s>": 29
+}

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 21.23,
+    "eval_loss": 0.84695965051651,
+    "eval_runtime": 90.2768,
+    "eval_samples": 2822,
+    "eval_samples_per_second": 31.259,
+    "eval_steps_per_second": 2.614,
+    "eval_wer": 0.18976083707025412,
+    "train_loss": 0.6778903991699219,
+    "train_runtime": 11598.0436,
+    "train_samples": 11291,
+    "train_samples_per_second": 20.693,
+    "train_steps_per_second": 0.862
+}

alphabet.json

@@ -0,0 +1 @@
+{"labels": [" ", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "\u2047", "", "<s>", "</s>"], "is_bpe": false}

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": true,
+  "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": 0.0,
+  "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": 12,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 12,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.075,
+  "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": 28,
+  "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": "4.24.0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 31,
+  "xvector_output_dim": 512
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 21.23,
+    "eval_loss": 0.84695965051651,
+    "eval_runtime": 90.2768,
+    "eval_samples": 2822,
+    "eval_samples_per_second": 31.259,
+    "eval_steps_per_second": 2.614,
+    "eval_wer": 0.18976083707025412
+}

language_model/attrs.json

@@ -0,0 +1 @@
+{"alpha": 0.5, "beta": 1.5, "unk_score_offset": -10.0, "score_boundary": true}

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": "Wav2Vec2ProcessorWithLM",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:845c59c44457c2c1fc9ff10df658c9cd1c8a4e1d577de922f373e90fb6378313
+size 1262028973

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 +1,11 @@
+{
+  "bos_token": "<s>",
+  "do_lower_case": false,
+  "eos_token": "</s>",
+  "pad_token": "[PAD]",
+  "processor_class": "Wav2Vec2ProcessorWithLM",
+  "replace_word_delimiter_char": " ",
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "unk_token": "[UNK]",
+  "word_delimiter_token": "|"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 21.23,
+    "train_loss": 0.6778903991699219,
+    "train_runtime": 11598.0436,
+    "train_samples": 11291,
+    "train_samples_per_second": 20.693,
+    "train_steps_per_second": 0.862
+}

trainer_state.json

@@ -0,0 +1,265 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 21.231422505307854,
+  "global_step": 10000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.06,
+      "eval_loss": 3.1697003841400146,
+      "eval_runtime": 96.1702,
+      "eval_samples_per_second": 29.344,
+      "eval_steps_per_second": 2.454,
+      "eval_wer": 1.0,
+      "step": 500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 9.98e-05,
+      "loss": 3.1489,
+      "step": 1000
+    },
+    {
+      "epoch": 2.12,
+      "eval_loss": 1.418416142463684,
+      "eval_runtime": 83.1708,
+      "eval_samples_per_second": 33.93,
+      "eval_steps_per_second": 2.838,
+      "eval_wer": 0.5677877428998506,
+      "step": 1000
+    },
+    {
+      "epoch": 3.18,
+      "eval_loss": 0.8497655987739563,
+      "eval_runtime": 83.5791,
+      "eval_samples_per_second": 33.764,
+      "eval_steps_per_second": 2.824,
+      "eval_wer": 0.3365844544095665,
+      "step": 1500
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 8.89111111111111e-05,
+      "loss": 0.8499,
+      "step": 2000
+    },
+    {
+      "epoch": 4.25,
+      "eval_loss": 0.8088647127151489,
+      "eval_runtime": 83.4386,
+      "eval_samples_per_second": 33.821,
+      "eval_steps_per_second": 2.828,
+      "eval_wer": 0.2755231689088191,
+      "step": 2000
+    },
+    {
+      "epoch": 5.31,
+      "eval_loss": 0.7338947057723999,
+      "eval_runtime": 83.1453,
+      "eval_samples_per_second": 33.941,
+      "eval_steps_per_second": 2.838,
+      "eval_wer": 0.29633781763826605,
+      "step": 2500
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 7.780000000000001e-05,
+      "loss": 0.5901,
+      "step": 3000
+    },
+    {
+      "epoch": 6.37,
+      "eval_loss": 0.6375774145126343,
+      "eval_runtime": 83.3907,
+      "eval_samples_per_second": 33.841,
+      "eval_steps_per_second": 2.83,
+      "eval_wer": 0.24024663677130045,
+      "step": 3000
+    },
+    {
+      "epoch": 7.43,
+      "eval_loss": 0.688980221748352,
+      "eval_runtime": 85.6719,
+      "eval_samples_per_second": 32.94,
+      "eval_steps_per_second": 2.755,
+      "eval_wer": 0.2335575485799701,
+      "step": 3500
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 6.66888888888889e-05,
+      "loss": 0.4724,
+      "step": 4000
+    },
+    {
+      "epoch": 8.49,
+      "eval_loss": 0.6844218969345093,
+      "eval_runtime": 88.5555,
+      "eval_samples_per_second": 31.867,
+      "eval_steps_per_second": 2.665,
+      "eval_wer": 0.22402840059790732,
+      "step": 4000
+    },
+    {
+      "epoch": 9.55,
+      "eval_loss": 0.6900164484977722,
+      "eval_runtime": 87.8731,
+      "eval_samples_per_second": 32.114,
+      "eval_steps_per_second": 2.686,
+      "eval_wer": 0.22223467862481316,
+      "step": 4500
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 5.5577777777777784e-05,
+      "loss": 0.3981,
+      "step": 5000
+    },
+    {
+      "epoch": 10.62,
+      "eval_loss": 0.7050736546516418,
+      "eval_runtime": 88.2857,
+      "eval_samples_per_second": 31.964,
+      "eval_steps_per_second": 2.673,
+      "eval_wer": 0.21229446935724963,
+      "step": 5000
+    },
+    {
+      "epoch": 11.68,
+      "eval_loss": 0.6670752167701721,
+      "eval_runtime": 88.0544,
+      "eval_samples_per_second": 32.048,
+      "eval_steps_per_second": 2.68,
+      "eval_wer": 0.2095291479820628,
+      "step": 5500
+    },
+    {
+      "epoch": 12.74,
+      "learning_rate": 4.4466666666666666e-05,
+      "loss": 0.3436,
+      "step": 6000
+    },
+    {
+      "epoch": 12.74,
+      "eval_loss": 0.7424847483634949,
+      "eval_runtime": 89.6214,
+      "eval_samples_per_second": 31.488,
+      "eval_steps_per_second": 2.633,
+      "eval_wer": 0.20489536621823617,
+      "step": 6000
+    },
+    {
+      "epoch": 13.8,
+      "eval_loss": 0.7134818434715271,
+      "eval_runtime": 96.8293,
+      "eval_samples_per_second": 29.144,
+      "eval_steps_per_second": 2.437,
+      "eval_wer": 0.19943946188340808,
+      "step": 6500
+    },
+    {
+      "epoch": 14.86,
+      "learning_rate": 3.3355555555555554e-05,
+      "loss": 0.2925,
+      "step": 7000
+    },
+    {
+      "epoch": 14.86,
+      "eval_loss": 0.7350390553474426,
+      "eval_runtime": 94.8583,
+      "eval_samples_per_second": 29.75,
+      "eval_steps_per_second": 2.488,
+      "eval_wer": 0.20123318385650224,
+      "step": 7000
+    },
+    {
+      "epoch": 15.92,
+      "eval_loss": 0.785513699054718,
+      "eval_runtime": 92.4034,
+      "eval_samples_per_second": 30.54,
+      "eval_steps_per_second": 2.554,
+      "eval_wer": 0.1945067264573991,
+      "step": 7500
+    },
+    {
+      "epoch": 16.99,
+      "learning_rate": 2.225555555555556e-05,
+      "loss": 0.2525,
+      "step": 8000
+    },
+    {
+      "epoch": 16.99,
+      "eval_loss": 0.7933125495910645,
+      "eval_runtime": 92.9055,
+      "eval_samples_per_second": 30.375,
+      "eval_steps_per_second": 2.54,
+      "eval_wer": 0.19461883408071748,
+      "step": 8000
+    },
+    {
+      "epoch": 18.05,
+      "eval_loss": 0.801628589630127,
+      "eval_runtime": 92.1414,
+      "eval_samples_per_second": 30.627,
+      "eval_steps_per_second": 2.561,
+      "eval_wer": 0.19151718983557547,
+      "step": 8500
+    },
+    {
+      "epoch": 19.11,
+      "learning_rate": 1.1155555555555556e-05,
+      "loss": 0.2285,
+      "step": 9000
+    },
+    {
+      "epoch": 19.11,
+      "eval_loss": 0.8283743262290955,
+      "eval_runtime": 94.3902,
+      "eval_samples_per_second": 29.897,
+      "eval_steps_per_second": 2.5,
+      "eval_wer": 0.1906576980568012,
+      "step": 9000
+    },
+    {
+      "epoch": 20.17,
+      "eval_loss": 0.8274713754653931,
+      "eval_runtime": 89.6678,
+      "eval_samples_per_second": 31.472,
+      "eval_steps_per_second": 2.632,
+      "eval_wer": 0.19024663677130044,
+      "step": 9500
+    },
+    {
+      "epoch": 21.23,
+      "learning_rate": 4.444444444444445e-08,
+      "loss": 0.2025,
+      "step": 10000
+    },
+    {
+      "epoch": 21.23,
+      "eval_loss": 0.84695965051651,
+      "eval_runtime": 89.481,
+      "eval_samples_per_second": 31.537,
+      "eval_steps_per_second": 2.637,
+      "eval_wer": 0.18976083707025412,
+      "step": 10000
+    },
+    {
+      "epoch": 21.23,
+      "step": 10000,
+      "total_flos": 2.495144460541511e+19,
+      "train_loss": 0.6778903991699219,
+      "train_runtime": 11598.0436,
+      "train_samples_per_second": 20.693,
+      "train_steps_per_second": 0.862
+    }
+  ],
+  "max_steps": 10000,
+  "num_train_epochs": 22,
+  "total_flos": 2.495144460541511e+19,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:141d34c93007923eeaf791c556f14f4f38e36e5f231ecdb30063a30935279018
+size 3707

vocab.json

@@ -0,0 +1,31 @@
+{
+  "[PAD]": 28,
+  "[UNK]": 27,
+  "a": 1,
+  "b": 2,
+  "c": 3,
+  "d": 4,
+  "e": 5,
+  "f": 6,
+  "g": 7,
+  "h": 8,
+  "i": 9,
+  "j": 10,
+  "k": 11,
+  "l": 12,
+  "m": 13,
+  "n": 14,
+  "o": 15,
+  "p": 16,
+  "q": 17,
+  "r": 18,
+  "s": 19,
+  "t": 20,
+  "u": 21,
+  "v": 22,
+  "w": 23,
+  "x": 24,
+  "y": 25,
+  "z": 26,
+  "|": 0
+}