xls-r-300m-toi-nst
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9700
- Wer: 0.8674
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.1632 | 0.3 | 500 | 0.4058 | 0.6069 |
| 0.7323 | 0.59 | 1000 | 0.2550 | 0.4421 |
| 0.6532 | 0.89 | 1500 | 0.2360 | 0.3812 |
| 0.592 | 1.18 | 2000 | 0.2007 | 0.3412 |
| 0.57 | 1.48 | 2500 | 0.1979 | 0.3382 |
| 0.5558 | 1.77 | 3000 | 0.1853 | 0.2995 |
| 0.5451 | 2.07 | 3500 | 0.1887 | 0.3151 |
| 0.5451 | 2.36 | 4000 | 0.5467 | 0.5383 |
| 1.3259 | 2.66 | 4500 | 0.9700 | 0.8674 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.10.0
- Tokenizers 0.13.2