<!-- 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. -->
wav2vec2-6
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 5.2459
- Wer: 1.0
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: 0.003
- train_batch_size: 32
- 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: 400
- num_epochs: 30
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
4.5873 | 1.56 | 200 | 5.4586 | 1.0 |
4.1846 | 3.12 | 400 | 5.2278 | 1.0 |
4.1711 | 4.69 | 600 | 5.3131 | 1.0 |
4.1581 | 6.25 | 800 | 5.2558 | 1.0 |
4.1275 | 7.81 | 1000 | 5.2556 | 1.0 |
4.1452 | 9.38 | 1200 | 5.2637 | 1.0 |
4.1614 | 10.94 | 1400 | 5.2847 | 1.0 |
4.1667 | 12.5 | 1600 | 5.2349 | 1.0 |
4.1471 | 14.06 | 1800 | 5.2850 | 1.0 |
4.1268 | 15.62 | 2000 | 5.2510 | 1.0 |
4.1701 | 17.19 | 2200 | 5.2605 | 1.0 |
4.1459 | 18.75 | 2400 | 5.2493 | 1.0 |
4.1411 | 20.31 | 2600 | 5.2649 | 1.0 |
4.1351 | 21.88 | 2800 | 5.2541 | 1.0 |
4.1442 | 23.44 | 3000 | 5.2459 | 1.0 |
4.1805 | 25.0 | 3200 | 5.2232 | 1.0 |
4.1262 | 26.56 | 3400 | 5.2384 | 1.0 |
4.145 | 28.12 | 3600 | 5.2522 | 1.0 |
4.142 | 29.69 | 3800 | 5.2459 | 1.0 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1