metadata
language:
- eng
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- Gabi00/english-mistakes
metrics:
- wer
model-index:
- name: Whisper Small Eng - Gabriel Mora
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: English-mistakes
type: Gabi00/english-mistakes
config: default
split: validation
args: 'config: eng, split: test'
metrics:
- name: Wer
type: wer
value: 13.110781686527167
Whisper Small Eng - Gabriel Mora
This model is a fine-tuned version of openai/whisper-small on the English-mistakes dataset. It achieves the following results on the evaluation set:
- Loss: 0.3784
- Wer: 13.1108
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: 28
- eval_batch_size: 28
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.6399 | 0.4444 | 500 | 0.4393 | 13.1649 |
| 0.526 | 0.8889 | 1000 | 0.4070 | 14.4792 |
| 0.4598 | 1.3333 | 1500 | 0.3948 | 14.2654 |
| 0.4094 | 1.7778 | 2000 | 0.3806 | 14.4050 |
| 0.3556 | 2.2222 | 2500 | 0.3816 | 13.2305 |
| 0.3304 | 2.6667 | 3000 | 0.3784 | 13.1108 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.1.0+cu118
- Datasets 2.20.0
- Tokenizers 0.19.1