Whisper Small Hi - Sanchit Gandhi
This model is a fine-tuned version of nurzhanit/whisper-enhanced-ml on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0002
- Wer: 0.0266
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: 50
- training_steps: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0357 | 0.8621 | 50 | 0.0117 | 1.5708 |
| 0.0132 | 1.7241 | 100 | 0.0055 | 0.5325 |
| 0.0068 | 2.5862 | 150 | 0.0033 | 0.3727 |
| 0.0022 | 3.4483 | 200 | 0.0012 | 0.1065 |
| 0.0004 | 4.3103 | 250 | 0.0005 | 0.0532 |
| 0.0008 | 5.1724 | 300 | 0.0003 | 0.0266 |
| 0.0002 | 6.0345 | 350 | 0.0003 | 0.0266 |
| 0.0004 | 6.8966 | 400 | 0.0003 | 0.0266 |
| 0.0001 | 7.7586 | 450 | 0.0003 | 0.0266 |
| 0.0002 | 8.6207 | 500 | 0.0002 | 0.0266 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.2
- Tokenizers 0.19.1
- Downloads last month
- 58,427
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.
Model tree for nurzhanit/whisper-omg-2
Unable to build the model tree, the base model loops to the model itself. Learn more.