metadata
base_model: openai/whisper-small
datasets:
- mozilla-foundation/common_voice_16_1
language:
- nan
library_name: peft
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: Whisper small Taiwanese - LoRA
results: []
Whisper small Taiwanese - LoRA
This model is a fine-tuned version of openai/whisper-small on the Common Voice 16.1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.9306
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: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 9
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.0057 | 1.0 | 2528 | 1.0098 |
| 0.846 | 2.0 | 5056 | 0.9359 |
| 0.7486 | 3.0 | 7585 | 0.8879 |
| 0.6056 | 4.0 | 10112 | 0.8839 |
| 0.4569 | 5.0 | 12640 | 0.9000 |
| 0.4929 | 6.0 | 15169 | 0.8907 |
| 0.3831 | 7.0 | 17696 | 0.9161 |
| 0.379 | 8.0 | 20225 | 0.9116 |
| 0.3041 | 9.0 | 22752 | 0.9306 |
Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.2.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2