metadata
language:
- eng
license: apache-2.0
base_model: openai/whisper-small
tags:
- generated_from_trainer
datasets:
- whisper/smith-and-brock
metrics:
- wer
model-index:
- name: Whisper Small Smith & Brock
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Whisper Smith & Brock
type: whisper/smith-and-brock
args: 'config: eng, split: validation'
metrics:
- name: Wer
type: wer
value: 98.0392156862745
Whisper Small Smith & Brock
This model is a fine-tuned version of openai/whisper-small on the Whisper Smith & Brock dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Wer: 98.0392
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0 | 1000.0 | 1000 | 0.0000 | 22.5490 |
| 0.0 | 2000.0 | 2000 | 0.0000 | 98.0392 |
| 0.0 | 3000.0 | 3000 | 0.0000 | 98.0392 |
| 0.0 | 4000.0 | 4000 | 0.0000 | 98.0392 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1