wav2vec2-xls-r-300m-swahili-cv-fleurs-alffa
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2092
- Wer: 0.0568
- Cer: 0.0478
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.5593 | 1.0 | 2007 | 0.3037 | 0.0790 | 0.0647 |
| 0.3965 | 2.0 | 4014 | 0.2326 | 0.0634 | 0.0531 |
| 0.2894 | 3.0 | 6021 | 0.2092 | 0.0568 | 0.0478 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1
- Datasets 2.18.1.dev0
- Tokenizers 0.15.2