wav2vec2-large-xlsr-53-indic_voices-61983-normalized-30_70-1e-4-steps-12000-FT
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6266
- Global Wer: 40.9022
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant_with_warmup
- lr_scheduler_warmup_steps: 500
- training_steps: 12000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Global Wer |
|---|---|---|---|---|
| 4.0551 | 0.5789 | 1000 | 1.4932 | 81.3472 |
| 2.2846 | 1.1575 | 2000 | 0.8803 | 57.7037 |
| 2.1005 | 1.7363 | 3000 | 0.7720 | 52.4675 |
| 1.7694 | 2.3149 | 4000 | 0.7188 | 48.9098 |
| 1.6905 | 2.8938 | 5000 | 0.7031 | 47.2425 |
| 1.4472 | 3.4724 | 6000 | 0.6424 | 43.8688 |
| 1.4250 | 4.0509 | 7000 | 0.6594 | 44.4321 |
| 1.3892 | 4.6298 | 8000 | 0.6371 | 43.2555 |
| 1.1820 | 5.2084 | 9000 | 0.6464 | 42.8595 |
| 1.2976 | 5.7873 | 10000 | 0.6190 | 42.1736 |
| 1.1530 | 6.3658 | 11000 | 0.6383 | 42.6253 |
| 1.1736 | 6.9447 | 12000 | 0.6266 | 40.9022 |
Framework versions
- Transformers 5.13.0
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.22.2