wav2vec2_base_1.5
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3070
- Wer: 0.2230
- Cer: 0.0794
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 |
---|---|---|---|---|---|
1.4392 | 1.0 | 500 | 0.6651 | 0.4226 | 0.1684 |
0.9721 | 2.0 | 1000 | 0.6476 | 0.4063 | 0.1641 |
0.7793 | 3.0 | 1500 | 0.5237 | 0.3547 | 0.1412 |
0.6246 | 4.0 | 2000 | 0.4705 | 0.3275 | 0.1271 |
0.5062 | 5.0 | 2500 | 0.4313 | 0.3013 | 0.1147 |
0.4084 | 6.0 | 3000 | 0.3873 | 0.2798 | 0.1059 |
0.324 | 7.0 | 3500 | 0.3632 | 0.2534 | 0.0945 |
0.2531 | 8.0 | 4000 | 0.3387 | 0.2362 | 0.0857 |
0.197 | 9.0 | 4500 | 0.3219 | 0.2267 | 0.0806 |
0.1605 | 10.0 | 5000 | 0.3070 | 0.2230 | 0.0794 |
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.