metadata
license: apache-2.0
base_model: openai/whisper-base.en
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: whisper-base.en-fsc
results: []
whisper-base.en-fsc
This model is a fine-tuned version of openai/whisper-base.en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.0278
- Accuracy: 0.5630
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-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 192
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.9972 | 263 | 3.7447 | 0.0962 |
| No log | 1.9981 | 527 | 2.8087 | 0.3060 |
| No log | 2.9991 | 791 | 2.3083 | 0.4062 |
| 2.9232 | 4.0 | 1055 | 2.0094 | 0.4940 |
| 2.9232 | 4.9972 | 1318 | 1.9099 | 0.5321 |
| 2.9232 | 5.9981 | 1582 | 1.9257 | 0.5479 |
| 2.9232 | 6.9991 | 1846 | 2.0132 | 0.5479 |
| 0.8199 | 8.0 | 2110 | 2.1486 | 0.5444 |
| 0.8199 | 8.9972 | 2373 | 2.2976 | 0.5440 |
| 0.8199 | 9.9981 | 2637 | 2.4131 | 0.5453 |
| 0.8199 | 10.9991 | 2901 | 2.5031 | 0.5523 |
| 0.1503 | 12.0 | 3165 | 2.6273 | 0.5544 |
| 0.1503 | 12.9972 | 3428 | 2.7233 | 0.5581 |
| 0.1503 | 13.9981 | 3692 | 2.8470 | 0.5498 |
| 0.1503 | 14.9991 | 3956 | 2.8848 | 0.5589 |
| 0.0246 | 16.0 | 4220 | 2.9497 | 0.5605 |
| 0.0246 | 16.9972 | 4483 | 2.9992 | 0.5612 |
| 0.0246 | 17.9981 | 4747 | 3.0278 | 0.5630 |
| 0.0043 | 18.9991 | 5011 | 3.0502 | 0.5629 |
| 0.0043 | 19.9431 | 5260 | 3.0561 | 0.5629 |
Framework versions
- Transformers 4.43.3
- Pytorch 2.2.2+cu121
- Datasets 2.18.0
- Tokenizers 0.19.1