metadata
language:
- ta
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- whisper-small-preon-test-1
metrics:
- wer
model-index:
- name: Whisper small
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: custom dataset
type: whisper-small-preon-test-1
metrics:
- name: Wer
type: wer
value: 11.920529801324504
Whisper small
This model is a fine-tuned version of openai/whisper-small on the custom dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.1046
- Wer Ortho: 11.8421
- Wer: 11.9205
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
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.4335 | 5.0 | 100 | 0.1326 | 11.8421 | 9.2715 |
| 0.0049 | 10.0 | 200 | 0.1332 | 15.7895 | 13.9073 |
| 0.0001 | 15.0 | 300 | 0.1019 | 11.8421 | 11.9205 |
| 0.0 | 20.0 | 400 | 0.1041 | 11.8421 | 11.9205 |
| 0.0 | 25.0 | 500 | 0.1046 | 11.8421 | 11.9205 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1