w2vbert-clean-silence-v2
This model is a fine-tuned version of GodwillN/w2vbert-waxal-corrected on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4585
- Wer: 0.2618
- Cer: 0.0846
- Combined Err: 0.1732
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: 5e-06
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.02
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer | Combined Err |
|---|---|---|---|---|---|---|
| 0.1955 | 1.0 | 2242 | 0.4585 | 0.2618 | 0.0846 | 0.1732 |
Framework versions
- Transformers 4.53.2
- Pytorch 2.12.0+cu130
- Datasets 3.6.0
- Tokenizers 0.21.4
- Downloads last month
- 17
Model tree for GodwillN/w2vbert-clean-silence-v2
Base model
GodwillN/w2vbert-waxal-corrected