metadata
license: apache-2.0
base_model: facebook/wav2vec2-xls-r-300m
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-large-xls-r-300m-Arabic-phoneme-based-MDD-experiment2
results: []
wav2vec2-large-xls-r-300m-Arabic-phoneme-based-MDD-experiment2
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0380
- Per: 0.0182
- Wer: 0.0209
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: 0.0005
- train_batch_size: 8
- eval_batch_size: 6
- 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: linear
- lr_scheduler_warmup_steps: 250
- num_epochs: 30.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Per | Wer |
|---|---|---|---|---|---|
| 2.3253 | 1.0 | 546 | 1.6463 | 0.9997 | 0.9990 |
| 1.7194 | 2.0 | 1093 | 1.4798 | 0.9993 | 0.9985 |
| 1.6318 | 3.0 | 1640 | 1.2181 | 0.9304 | 0.9399 |
| 1.4017 | 4.0 | 2187 | 0.5801 | 0.2704 | 0.3889 |
| 0.9203 | 5.0 | 2733 | 0.1995 | 0.1330 | 0.1439 |
| 0.669 | 6.0 | 3280 | 0.1462 | 0.0867 | 0.0928 |
| 0.5585 | 7.0 | 3827 | 0.0925 | 0.0739 | 0.0752 |
| 0.4954 | 8.0 | 4374 | 0.0795 | 0.0601 | 0.0573 |
| 0.4536 | 9.0 | 4920 | 0.0714 | 0.0498 | 0.0530 |
| 0.4183 | 10.0 | 5467 | 0.0726 | 0.0652 | 0.0679 |
| 0.3896 | 11.0 | 6014 | 0.0782 | 0.0566 | 0.0553 |
| 0.3643 | 12.0 | 6561 | 0.0713 | 0.0502 | 0.0514 |
| 0.3391 | 13.0 | 7107 | 0.0928 | 0.0693 | 0.0702 |
| 0.3229 | 14.0 | 7654 | 0.0596 | 0.0503 | 0.0484 |
| 0.3038 | 15.0 | 8201 | 0.0651 | 0.0531 | 0.0547 |
| 0.2853 | 16.0 | 8748 | 0.0577 | 0.0483 | 0.0473 |
| 0.2736 | 17.0 | 9294 | 0.0542 | 0.0463 | 0.0485 |
| 0.2603 | 18.0 | 9841 | 0.0519 | 0.0327 | 0.0325 |
| 0.2472 | 19.0 | 10388 | 0.0513 | 0.0336 | 0.0351 |
| 0.2346 | 20.0 | 10935 | 0.0589 | 0.0281 | 0.0310 |
| 0.2265 | 21.0 | 11481 | 0.0499 | 0.0285 | 0.0312 |
| 0.214 | 22.0 | 12028 | 0.0485 | 0.0281 | 0.0342 |
| 0.2045 | 23.0 | 12575 | 0.0479 | 0.0273 | 0.0295 |
| 0.1917 | 24.0 | 13122 | 0.0457 | 0.0267 | 0.0286 |
| 0.1814 | 25.0 | 13668 | 0.0423 | 0.0224 | 0.0265 |
| 0.1769 | 26.0 | 14215 | 0.0442 | 0.0243 | 0.0264 |
| 0.1682 | 27.0 | 14762 | 0.0396 | 0.0200 | 0.0235 |
| 0.161 | 28.0 | 15309 | 0.0386 | 0.0190 | 0.0225 |
| 0.1572 | 29.0 | 15855 | 0.0383 | 0.0186 | 0.0216 |
| 0.1502 | 29.96 | 16380 | 0.0380 | 0.0182 | 0.0209 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu118
- Datasets 1.18.3
- Tokenizers 0.13.3