wav2vec2-large-xlsr-53-demo-colab
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:
- Loss: 1.4640
- Wer: 0.7304
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 |
|---|---|---|---|---|
| 17.8973 | 6.5620 | 400 | 3.4546 | 1.0 |
| 4.764 | 13.1157 | 800 | 1.4512 | 0.8837 |
| 0.7421 | 19.6777 | 1200 | 1.3554 | 0.7489 |
| 0.3618 | 26.2314 | 1600 | 1.4640 | 0.7304 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0