metadata
base_model: openai/whisper-tiny
language:
- it
library_name: transformers
license: apache-2.0
metrics:
- wer
tags:
- hf-asr-leaderboard
- generated_from_trainer
model-index:
- name: Whisper Tiny Italian Combine 5k - Chee Li
results: []
Whisper Tiny Italian Combine 5k - Chee Li
This model is a fine-tuned version of openai/whisper-tiny on the Google Fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.4933
- Wer: 52.2594
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 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 |
|---|---|---|---|---|
| 0.5398 | 0.0849 | 1000 | 0.6209 | 60.9740 |
| 0.4894 | 0.1699 | 2000 | 0.5541 | 56.0544 |
| 0.4558 | 0.2548 | 3000 | 0.5213 | 54.6387 |
| 0.4267 | 0.3398 | 4000 | 0.5010 | 52.4281 |
| 0.4225 | 0.4247 | 5000 | 0.4933 | 52.2594 |
Framework versions
- Transformers 4.46.2
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.20.1