Whisper Small Enlgish Common Voice
This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_11_0 en dataset. It achieves the following results on the evaluation set:
- Loss: 0.3261
- Wer: 12.9975
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: 64
- eval_batch_size: 32
- 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2259 | 0.2 | 1000 | 0.3857 | 15.1528 |
| 0.2584 | 0.4 | 2000 | 0.3618 | 14.5494 |
| 0.3919 | 0.6 | 3000 | 0.3964 | 17.4630 |
| 0.2566 | 0.8 | 4000 | 0.3349 | 13.4222 |
| 0.2232 | 1.0 | 5000 | 0.3261 | 12.9975 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.8.1.dev0
- Tokenizers 0.12.1
- Downloads last month
- 1
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Dataset used to train rishabhjain16/whisper-small-en
Evaluation results
- Wer on mozilla-foundation/common_voice_11_0 entest set self-reported12.997