metadata
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-large-asr-th-2
results: []
wav2vec2-large-asr-th-2
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2713
- Wer: 0.3245
- Cer: 0.0892
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: 2e-05
- train_batch_size: 12
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 3
- total_train_batch_size: 36
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.0865 | 0.18 | 1000 | 0.5156 | 0.3249 | 0.0892 |
| 0.1075 | 0.36 | 2000 | 0.5144 | 0.3265 | 0.0899 |
| 0.1973 | 0.53 | 3000 | 0.4405 | 0.3298 | 0.0905 |
| 0.8348 | 0.71 | 4000 | 0.2713 | 0.3245 | 0.0892 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3