wav2vec-XLSR-swahili
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2337
- Wer: 0.1816
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: 0.0003
- 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_steps: 500
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.463 | 0.87 | 1000 | 0.3897 | 0.3611 |
| 0.2431 | 1.74 | 2000 | 0.3008 | 0.2556 |
| 0.1731 | 2.61 | 3000 | 0.2726 | 0.2204 |
| 0.1311 | 3.47 | 4000 | 0.2527 | 0.2002 |
| 0.11 | 4.34 | 5000 | 0.2337 | 0.1816 |
Framework versions
- Transformers 4.33.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.14.3
- Tokenizers 0.13.3