<!-- 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-demo-M03-2
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2042
- Wer: 0.4965
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: 1000
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 22.0914 | 0.92 | 500 | 3.4247 | 1.0 |
| 3.3901 | 1.85 | 1000 | 2.9866 | 1.0 |
| 2.8827 | 2.77 | 1500 | 2.7737 | 1.0 |
| 2.5545 | 3.69 | 2000 | 2.0033 | 1.0 |
| 1.8375 | 4.61 | 2500 | 0.9457 | 1.2411 |
| 1.3094 | 5.54 | 3000 | 0.6196 | 1.0427 |
| 1.0079 | 6.46 | 3500 | 0.4693 | 0.9623 |
| 0.8174 | 7.38 | 4000 | 0.3809 | 0.8725 |
| 0.7038 | 8.3 | 4500 | 0.3147 | 0.7991 |
| 0.5961 | 9.23 | 5000 | 0.2672 | 0.7475 |
| 0.5066 | 10.15 | 5500 | 0.2440 | 0.7505 |
| 0.4589 | 11.07 | 6000 | 0.2259 | 0.6979 |
| 0.4379 | 11.99 | 6500 | 0.2133 | 0.7138 |
| 0.3835 | 12.92 | 7000 | 0.2315 | 0.6791 |
| 0.3431 | 13.84 | 7500 | 0.1889 | 0.6200 |
| 0.3273 | 14.76 | 8000 | 0.2021 | 0.6136 |
| 0.3087 | 15.68 | 8500 | 0.2159 | 0.5967 |
| 0.291 | 16.61 | 9000 | 0.2294 | 0.5828 |
| 0.2662 | 17.53 | 9500 | 0.1736 | 0.5675 |
| 0.2558 | 18.45 | 10000 | 0.1957 | 0.5531 |
| 0.2388 | 19.37 | 10500 | 0.2047 | 0.5372 |
| 0.2219 | 20.3 | 11000 | 0.1843 | 0.5208 |
| 0.2024 | 21.22 | 11500 | 0.2039 | 0.5278 |
| 0.187 | 22.14 | 12000 | 0.1810 | 0.5159 |
| 0.1937 | 23.06 | 12500 | 0.1856 | 0.5188 |
| 0.1792 | 23.99 | 13000 | 0.1769 | 0.5139 |
| 0.1642 | 24.91 | 13500 | 0.1994 | 0.5089 |
| 0.1665 | 25.83 | 14000 | 0.1911 | 0.5124 |
| 0.1589 | 26.75 | 14500 | 0.2000 | 0.5074 |
| 0.1599 | 27.68 | 15000 | 0.1912 | 0.4970 |
| 0.135 | 28.6 | 15500 | 0.2028 | 0.4936 |
| 0.1383 | 29.52 | 16000 | 0.2042 | 0.4965 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 1.18.3
- Tokenizers 0.13.2
