multilingual-xlsr-53-F1-new-CH
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: 0.8233
- Accuracy: 0.7886
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 9
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.1411 | 0.5714 | 400 | 1.0007 | 0.5850 |
| 1.0285 | 1.1429 | 800 | 0.9395 | 0.6243 |
| 0.9267 | 1.7143 | 1200 | 0.9307 | 0.6586 |
| 1.3018 | 2.2857 | 1600 | 0.6965 | 0.7414 |
| 0.4333 | 2.8571 | 2000 | 0.7846 | 0.7186 |
| 0.5758 | 3.4286 | 2400 | 0.7626 | 0.7407 |
| 0.5744 | 4.0 | 2800 | 0.6857 | 0.7721 |
| 0.7025 | 4.5714 | 3200 | 0.7270 | 0.7650 |
| 0.7066 | 5.1429 | 3600 | 0.7583 | 0.7679 |
| 0.4054 | 5.7143 | 4000 | 0.8564 | 0.7707 |
| 0.1102 | 6.2857 | 4400 | 0.6793 | 0.8007 |
| 0.427 | 6.8571 | 4800 | 0.7864 | 0.7793 |
| 1.092 | 7.4286 | 5200 | 0.7348 | 0.8000 |
| 0.0455 | 8.0 | 5600 | 0.7501 | 0.8014 |
| 0.9213 | 8.5714 | 6000 | 0.8233 | 0.7886 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.0