finetuning-whisper-tiny-swahili-asr-model
This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9872
- Wer: 65.2972
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.0941 | 3.6232 | 1000 | 0.9996 | 66.5619 |
| 0.8663 | 7.2464 | 2000 | 0.9685 | 64.9370 |
| 0.7385 | 10.8696 | 3000 | 0.9771 | 66.7650 |
| 0.6787 | 14.4928 | 4000 | 0.9872 | 65.2972 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.0+cu128
- Datasets 3.6.0
- Tokenizers 0.22.1