wav2vec2_base_1.5.3
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2868
- Wer: 0.2168
- Cer: 0.0758
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: 0.0003
- train_batch_size: 16
- 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: 500
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
---|---|---|---|---|---|
0.7186 | 0.89 | 500 | 0.6028 | 0.4056 | 0.1610 |
0.9547 | 1.78 | 1000 | 0.6175 | 0.4180 | 0.1697 |
0.8084 | 2.66 | 1500 | 0.4960 | 0.3489 | 0.1377 |
0.6525 | 3.55 | 2000 | 0.5084 | 0.3329 | 0.1323 |
0.548 | 4.44 | 2500 | 0.4705 | 0.2990 | 0.1166 |
0.4519 | 5.33 | 3000 | 0.4278 | 0.2805 | 0.1084 |
0.3772 | 6.22 | 3500 | 0.3823 | 0.2698 | 0.1019 |
0.3045 | 7.1 | 4000 | 0.3604 | 0.2528 | 0.0910 |
0.243 | 7.99 | 4500 | 0.3134 | 0.2345 | 0.0829 |
0.1969 | 8.88 | 5000 | 0.2982 | 0.2252 | 0.0808 |
0.1612 | 9.77 | 5500 | 0.2868 | 0.2168 | 0.0758 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 0
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.