temp
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: 0.4645
- Wer: 0.3527
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: 100
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4324 | 0.4 | 50 | 0.5800 | 0.4458 |
| 0.4027 | 0.8 | 100 | 0.5374 | 0.4109 |
| 0.3163 | 1.2 | 150 | 0.5285 | 0.3881 |
| 0.3064 | 1.6 | 200 | 0.5161 | 0.3815 |
| 0.3235 | 2.0 | 250 | 0.4645 | 0.3527 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu111
- Datasets 1.13.3
- Tokenizers 0.10.3