whisper-tiny-finetuned-common_voice
This model is a fine-tuned version of openai/whisper-tiny on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0201
- Accuracy: 0.9975
- F1: 0.9975
- Recall: 0.9975
- Precision: 0.9975
- Mcc: 0.9969
- Auc: 1.0000
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: 3e-05
- train_batch_size: 64
- eval_batch_size: 64
- 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: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision | Mcc | Auc |
---|---|---|---|---|---|---|---|---|---|
0.0044 | 1.0 | 25 | 0.0086 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0 |
0.0199 | 2.0 | 50 | 0.0015 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 |
0.0017 | 3.0 | 75 | 0.0342 | 0.9925 | 0.9925 | 0.9925 | 0.9926 | 0.9906 | 0.9999 |
0.0008 | 4.0 | 100 | 0.0038 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0 |
0.0006 | 5.0 | 125 | 0.0199 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0000 |
0.0004 | 6.0 | 150 | 0.0202 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0000 |
0.0004 | 7.0 | 175 | 0.0201 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0000 |
0.0003 | 8.0 | 200 | 0.0201 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0000 |
0.0003 | 9.0 | 225 | 0.0201 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0000 |
0.0003 | 10.0 | 250 | 0.0201 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0000 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 5