metadata
base_model: openai/whisper-large-v2
datasets:
- common_voice_16_1
library_name: peft
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: whisper-large-v2-ft-common_voice_16_1-241026-v1
results: []
whisper-large-v2-ft-common_voice_16_1-241026-v1
This model is a fine-tuned version of openai/whisper-large-v2 on the common_voice_16_1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1253
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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.2703 | 1.0 | 56 | 2.2818 |
| 1.602 | 2.0 | 112 | 0.6618 |
| 0.3079 | 3.0 | 168 | 0.1281 |
| 0.1397 | 4.0 | 224 | 0.1231 |
| 0.1213 | 5.0 | 280 | 0.1214 |
| 0.1084 | 6.0 | 336 | 0.1225 |
| 0.0983 | 7.0 | 392 | 0.1239 |
| 0.0918 | 8.0 | 448 | 0.1237 |
| 0.0862 | 9.0 | 504 | 0.1248 |
| 0.0824 | 10.0 | 560 | 0.1253 |
Framework versions
- PEFT 0.13.0
- Transformers 4.45.1
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.0