Wav2vec2-large-robust-Pronounciation-Evaluation
This model is a fine-tuned version of facebook/wav2vec2-large-robust on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7846
- Accuracy: 0.72
- F1: 0.72
- Precision: 0.72
- Recall: 0.72
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 100
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.7468 | 1.0 | 500 | 0.9762 | 0.616 | 0.616 | 0.616 | 0.616 |
| 0.492 | 2.0 | 1000 | 1.1308 | 0.536 | 0.536 | 0.536 | 0.536 |
| 0.619 | 3.0 | 1500 | 0.7913 | 0.688 | 0.688 | 0.688 | 0.688 |
| 0.56 | 4.0 | 2000 | 0.8142 | 0.67 | 0.67 | 0.67 | 0.67 |
| 0.4561 | 5.0 | 2500 | 0.7452 | 0.708 | 0.708 | 0.708 | 0.708 |
| 0.5474 | 6.0 | 3000 | 0.7846 | 0.72 | 0.72 | 0.72 | 0.72 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3