wav2vec2-large-xls-r-300m-turkish-colab
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice_6_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5881
- Wer: 0.4890
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: Use OptimizerNames.ADAMW_TORCH 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: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.9848 | 6.9043 | 400 | 0.8200 | 0.8013 |
| 0.3716 | 13.8 | 800 | 0.6589 | 0.6034 |
| 0.1546 | 20.6957 | 1200 | 0.5937 | 0.5260 |
| 0.0916 | 27.5913 | 1600 | 0.5881 | 0.4890 |
Framework versions
- Transformers 4.52.4
- Pytorch 2.6.0+cu124
- Datasets 2.18.0
- Tokenizers 0.21.2