<!-- 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-base-random-stopvoicing-1
This model is a fine-tuned version of on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3669
- Accuracy: 0.8702
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
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 24
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6871 | 0.99 | 20 | 0.6530 | 0.6213 |
| 0.6757 | 1.98 | 40 | 0.6507 | 0.6104 |
| 0.6193 | 2.96 | 60 | 0.4827 | 0.7691 |
| 0.5511 | 4.0 | 81 | 0.4494 | 0.7950 |
| 0.5076 | 4.99 | 101 | 0.4027 | 0.8283 |
| 0.4882 | 5.98 | 121 | 0.5145 | 0.7813 |
| 0.4728 | 6.96 | 141 | 0.4394 | 0.8120 |
| 0.4351 | 8.0 | 162 | 0.4163 | 0.8270 |
| 0.4432 | 8.99 | 182 | 0.3823 | 0.8392 |
| 0.4165 | 9.98 | 202 | 0.4307 | 0.8263 |
| 0.3947 | 10.96 | 222 | 0.3569 | 0.8604 |
| 0.4186 | 12.0 | 243 | 0.4431 | 0.8283 |
| 0.3948 | 12.99 | 263 | 0.3836 | 0.8522 |
| 0.3627 | 13.98 | 283 | 0.3778 | 0.8569 |
| 0.3922 | 14.96 | 303 | 0.3523 | 0.8624 |
| 0.3668 | 16.0 | 324 | 0.3543 | 0.8631 |
| 0.3676 | 16.99 | 344 | 0.3485 | 0.8610 |
| 0.3118 | 17.98 | 364 | 0.3838 | 0.8638 |
| 0.328 | 18.96 | 384 | 0.3509 | 0.8685 |
| 0.3387 | 20.0 | 405 | 0.3593 | 0.8685 |
| 0.3088 | 20.99 | 425 | 0.3596 | 0.8631 |
| 0.2942 | 21.98 | 445 | 0.3585 | 0.8713 |
| 0.3027 | 22.96 | 465 | 0.3644 | 0.8651 |
| 0.2913 | 23.7 | 480 | 0.3575 | 0.8692 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.0
- Datasets 2.7.1
- Tokenizers 0.13.2
