<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
wav2vec2-large-xlsr-53-english-pronunciation-evaluation-bs-v2
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: 0.8697
- Accuracy: 0.78
- F1: 0.7738
- Precision: 0.7735
- Recall: 0.78
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 |
---|---|---|---|---|---|---|---|
1.0774 | 1.0 | 500 | 0.9199 | 0.57 | 0.5728 | 0.6154 | 0.57 |
0.6526 | 2.0 | 1000 | 0.6857 | 0.7 | 0.6925 | 0.7167 | 0.7 |
0.3767 | 3.0 | 1500 | 0.5830 | 0.79 | 0.7887 | 0.7884 | 0.79 |
0.242 | 4.0 | 2000 | 0.7786 | 0.82 | 0.8160 | 0.8163 | 0.82 |
0.2691 | 5.0 | 2500 | 0.8399 | 0.814 | 0.8113 | 0.8109 | 0.814 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3