wav2vec2-large-xlsr-53-english-pronunciation-evaluation-aod-cut-balance
This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-xlsr-53-english on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0674
- Accuracy: 0.6055
- F1: 0.6017
- Precision: 0.6074
- Recall: 0.6055
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
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 1.0011 | 1.0 | 105 | 1.0494 | 0.5 | 0.4111 | 0.4721 | 0.5 |
| 0.7777 | 2.0 | 210 | 0.9454 | 0.5576 | 0.5178 | 0.5332 | 0.5576 |
| 0.7462 | 3.0 | 315 | 1.1190 | 0.5815 | 0.5649 | 0.5757 | 0.5815 |
| 0.6099 | 4.0 | 420 | 1.0299 | 0.6043 | 0.5975 | 0.5992 | 0.6043 |
| 0.4457 | 5.0 | 525 | 1.0674 | 0.6055 | 0.6017 | 0.6074 | 0.6055 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3