<!-- 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-base-timit-demo-colab57
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: 0.7328
- Wer: 0.4593
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: 60
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 4.9876 | 7.04 | 500 | 3.1483 | 1.0 |
| 1.4621 | 14.08 | 1000 | 0.6960 | 0.6037 |
| 0.4404 | 21.13 | 1500 | 0.6392 | 0.5630 |
| 0.2499 | 28.17 | 2000 | 0.6738 | 0.5281 |
| 0.1732 | 35.21 | 2500 | 0.6789 | 0.4952 |
| 0.1347 | 42.25 | 3000 | 0.7328 | 0.4835 |
| 0.1044 | 49.3 | 3500 | 0.7258 | 0.4840 |
| 0.0896 | 56.34 | 4000 | 0.7328 | 0.4593 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.11.0+cu113
- Datasets 1.18.3
- Tokenizers 0.10.3
