Wav2vec2-large-53-arabic_20_02
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the common_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.2812
- Wer: 24.1131
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- 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: 500
- training_steps: 6000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.6173 | 0.83 | 1000 | 0.5035 | 48.3239 |
| 0.4084 | 1.66 | 2000 | 0.3713 | 35.5494 |
| 0.2576 | 2.5 | 3000 | 0.3309 | 30.2076 |
| 0.2108 | 3.33 | 4000 | 0.3093 | 27.4785 |
| 0.1531 | 4.16 | 5000 | 0.2980 | 25.3745 |
| 0.1426 | 4.99 | 6000 | 0.2812 | 24.1131 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.10.0+cu113
- Datasets 2.1.0
- Tokenizers 0.12.1