<!-- 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. -->
This model was trained from scratch on the librispeech_asr dataset. It achieves the following results on the evaluation set:
- Loss: 5.2453
- Wer: 1.9070
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: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.6407 | 0.28 | 500 | 6.3882 | 1.9939 |
| 5.6351 | 0.56 | 1000 | 6.1492 | 2.0010 |
| 5.4896 | 0.84 | 1500 | 5.8014 | 1.9549 |
| 4.9941 | 1.12 | 2000 | 5.7117 | 1.9888 |
| 5.1524 | 1.4 | 2500 | 5.5260 | 1.3353 |
| 4.8291 | 1.68 | 3000 | 5.4030 | 1.9694 |
| 4.6539 | 1.96 | 3500 | 5.3507 | 1.9606 |
| 4.351 | 2.24 | 4000 | 5.4178 | 1.9678 |
| 4.1189 | 2.52 | 4500 | 5.2799 | 1.9154 |
| 3.9921 | 2.8 | 5000 | 5.2453 | 1.9070 |
Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu113
- Datasets 1.18.3
- Tokenizers 0.11.0
