wav2vec2-xlsr-53-espeak-cv-ft-evn4-ntsema-colab
This model is a fine-tuned version of ntsema/wav2vec2-xlsr-53-espeak-cv-ft-sah2-ntsema-colab on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 2.0821
- Wer: 0.9833
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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- 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
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 4.3115 | 6.15 | 400 | 1.6416 | 0.9867 |
| 0.9147 | 12.3 | 800 | 1.6538 | 0.9867 |
| 0.5301 | 18.46 | 1200 | 1.8461 | 0.98 |
| 0.2865 | 24.61 | 1600 | 2.0821 | 0.9833 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2