metadata
library_name: transformers
license: mit
base_model: openai/whisper-large-v3-turbo
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-large-v3-turbo-ft-cv-cy-en
results: []
whisper-large-v3-turbo-ft-cv-cy-en
This model is a fine-tuned version of openai/whisper-large-v3-turbo on the DewiBrynJones/commonvoice_18_0_cy_en train main dataset. It achieves the following results on the evaluation set:
- Loss: 0.2927
- Wer: 0.1577
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: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.6485 | 0.7075 | 1000 | 0.3581 | 0.2210 |
| 0.3362 | 1.4149 | 2000 | 0.3094 | 0.1831 |
| 0.1504 | 2.1224 | 3000 | 0.2957 | 0.1699 |
| 0.1558 | 2.8299 | 4000 | 0.2816 | 0.1646 |
| 0.0619 | 3.5373 | 5000 | 0.2927 | 0.1577 |
Framework versions
- Transformers 4.46.1
- Pytorch 2.5.1+cu124
- Datasets 3.0.2
- Tokenizers 0.20.1