torgo-sentences-unlimited
This model is a fine-tuned version of yongjian/wav2vec2-large-a on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4841
- Wer: 0.2696
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: 4
- 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 |
|---|---|---|---|---|
| 19.2722 | 3.65 | 500 | 5.3840 | 0.9950 |
| 3.5604 | 7.3 | 1000 | 3.0665 | 1.0 |
| 2.9063 | 10.95 | 1500 | 2.2699 | 1.0020 |
| 1.1228 | 14.6 | 2000 | 0.5434 | 0.3637 |
| 0.4145 | 18.25 | 2500 | 0.5388 | 0.2934 |
| 0.2217 | 21.9 | 3000 | 0.4926 | 0.2874 |
| 0.1832 | 25.55 | 3500 | 0.4761 | 0.2785 |
| 0.145 | 29.2 | 4000 | 0.4841 | 0.2696 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 1.18.3
- Tokenizers 0.13.2