wav2vec2-xls-r-300m-hy-AM-CV8-v1
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.9145
- Wer: 0.9598
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 300
- num_epochs: 170
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.7132 | 83.31 | 500 | 1.9274 | 1.0523 |
| 1.017 | 166.62 | 1000 | 0.9145 | 0.9598 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu111
- Datasets 1.18.3
- Tokenizers 0.10.3