wav2vec2-XLS-R-Fleurs-demo-google-colab-Ezra_William_Prod6
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the xtreme_s dataset. It achieves the following results on the evaluation set:
- Loss: 0.9273
- Wer: 0.4653
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: 0.001
- 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: 600
- num_epochs: 120
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 4.8946 | 18.18 | 300 | 2.8697 | 1.0 |
| 1.0963 | 36.36 | 600 | 0.8022 | 0.5904 |
| 0.1715 | 54.55 | 900 | 0.9387 | 0.5567 |
| 0.0953 | 72.73 | 1200 | 0.9120 | 0.5082 |
| 0.0651 | 90.91 | 1500 | 0.9362 | 0.4854 |
| 0.044 | 109.09 | 1800 | 0.9273 | 0.4653 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1