Whisper Base Ru - MonsterVen
This model is a fine-tuned version of openai/whisper-small on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2071
- Wer: 16.1078
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- 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: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2114 | 0.6064 | 1000 | 0.2317 | 19.0741 |
| 0.0918 | 1.2129 | 2000 | 0.2124 | 17.1833 |
| 0.0887 | 1.8193 | 3000 | 0.2045 | 16.3947 |
| 0.0449 | 2.4257 | 4000 | 0.2071 | 16.1078 |
Framework versions
- Transformers 4.52.4
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1
- Downloads last month
- 3
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for MonsterVen/whisper-base-ru
Base model
openai/whisper-small