Incremental Swahili Luganda
This model is a fine-tuned version of openai/whisper-small on the Mix data dataset. It achieves the following results on the evaluation set:
- Loss: 0.3827
- Wer: 36.1778
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: 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 |
|---|---|---|---|---|
| 0.3394 | 0.0740 | 500 | 0.4718 | 44.4111 |
| 0.3858 | 0.1481 | 1000 | 0.4562 | 42.6277 |
| 0.3431 | 0.2221 | 1500 | 0.4337 | 40.9521 |
| 0.3576 | 0.2962 | 2000 | 0.4183 | 39.3431 |
| 0.3434 | 0.3702 | 2500 | 0.4046 | 38.4007 |
| 0.3242 | 0.4443 | 3000 | 0.3948 | 37.2575 |
| 0.3604 | 0.5183 | 3500 | 0.3871 | 36.8201 |
| 0.2958 | 0.5924 | 4000 | 0.3827 | 36.1778 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.2+cu118
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 4