wav2vec2-base-demo-colab
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: 0.4171
- Wer: 0.3452
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: 8
- 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: 30
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
3.0054 | 4.0 | 500 | 1.5456 | 0.9005 |
0.8183 | 8.0 | 1000 | 0.4738 | 0.4839 |
0.3019 | 12.0 | 1500 | 0.4280 | 0.4047 |
0.1738 | 16.0 | 2000 | 0.4584 | 0.3738 |
0.1285 | 20.0 | 2500 | 0.4418 | 0.3593 |
0.1104 | 24.0 | 3000 | 0.4110 | 0.3479 |
0.0828 | 28.0 | 3500 | 0.4171 | 0.3452 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu102
- Datasets 1.14.0
- Tokenizers 0.10.3
- 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.