metadata
library_name: peft
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-model-small-ro-finetune-5k-15-15
results: []
whisper-model-small-ro-finetune-5k-15-15
This model is a fine-tuned version of openai/whisper-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0961
- Wer: 0.7677
- Cer: 0.5417
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: 32
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- 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: 30
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 3.6558 | 1.0 | 82 | 1.7449 | 0.8961 | 0.5965 |
| 1.5043 | 2.0 | 164 | 1.2044 | 1.0445 | 1.1512 |
| 1.3584 | 3.0 | 246 | 1.1374 | 0.9 | 0.8215 |
| 1.2921 | 4.0 | 328 | 1.1058 | 0.7583 | 0.6019 |
| 1.2643 | 5.0 | 410 | 1.0961 | 0.7677 | 0.5417 |
Framework versions
- PEFT 0.15.2
- Transformers 4.52.3
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1