Whisper Small Maori
This model is a fine-tuned version of openai/whisper-small on the google/fleurs mi_nz dataset. It achieves the following results on the evaluation set:
- Loss: 0.7756
- Wer: 30.4816
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: 100
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2693 | 7.02 | 100 | 0.6741 | 35.4845 |
| 0.0084 | 15.01 | 200 | 0.7756 | 30.4816 |
| 0.0029 | 23.0 | 300 | 0.8154 | 31.4744 |
| 0.002 | 30.02 | 400 | 0.8320 | 31.3777 |
| 0.0017 | 38.01 | 500 | 0.8372 | 31.5163 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2