<!-- 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: 0.6977
- Wer: 0.1231
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 25.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 6.1467 | 1.68 | 1500 | 6.0558 | 1.3243 |
| 5.4388 | 3.36 | 3000 | 5.4711 | 1.5604 |
| 3.3434 | 5.04 | 4500 | 3.4808 | 0.7461 |
| 1.5259 | 6.73 | 6000 | 2.1931 | 0.3430 |
| 1.4285 | 8.41 | 7500 | 1.5883 | 0.2784 |
| 1.0687 | 10.09 | 9000 | 1.2481 | 0.2069 |
| 0.6425 | 11.77 | 10500 | 1.0507 | 0.1758 |
| 0.7147 | 13.45 | 12000 | 0.9397 | 0.1584 |
| 0.5083 | 15.13 | 13500 | 0.8452 | 0.1453 |
| 0.4287 | 16.82 | 15000 | 0.7915 | 0.1388 |
| 0.3499 | 18.5 | 16500 | 0.7477 | 0.1315 |
| 0.3733 | 20.18 | 18000 | 0.7307 | 0.1287 |
| 0.2609 | 21.86 | 19500 | 0.7061 | 0.1263 |
| 0.2602 | 23.54 | 21000 | 0.6977 | 0.1231 |
Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu113
- Datasets 1.18.3
- Tokenizers 0.11.0
