wav2vec2-dstc2
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: inf
- Wer: 0.3375
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: 32
- eval_batch_size: 1
- 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: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.3005 | 1.0 | 488 | inf | 1.0 |
| 2.1918 | 2.0 | 976 | inf | 0.7492 |
| 1.0499 | 3.0 | 1464 | inf | 0.4338 |
| 0.9099 | 4.0 | 1952 | inf | 0.4005 |
| 0.6575 | 5.0 | 2440 | inf | 0.3900 |
| 0.6008 | 6.0 | 2928 | inf | 0.3705 |
| 0.5742 | 7.0 | 3416 | inf | 0.3916 |
| 0.57 | 8.0 | 3904 | inf | 0.3573 |
| 0.4874 | 9.0 | 4392 | inf | 0.3526 |
| 0.4304 | 10.0 | 4880 | inf | 0.3609 |
| 0.4189 | 11.0 | 5368 | inf | 0.3558 |
| 0.4532 | 12.0 | 5856 | inf | 0.3503 |
| 0.4088 | 13.0 | 6344 | inf | 0.3454 |
| 0.3515 | 14.0 | 6832 | inf | 0.3437 |
| 0.3596 | 15.0 | 7320 | inf | 0.3411 |
| 0.3299 | 16.0 | 7808 | inf | 0.3393 |
| 0.2859 | 17.0 | 8296 | inf | 0.3397 |
| 0.2546 | 18.0 | 8784 | inf | 0.3379 |
| 0.2632 | 19.0 | 9272 | inf | 0.3369 |
| 0.282 | 20.0 | 9760 | inf | 0.3375 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 0