<!-- 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-demo-F04-2
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3203
- Wer: 0.5353
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_steps: 1000
- num_epochs: 30
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
23.5576 | 0.89 | 500 | 3.3654 | 1.0 |
3.3953 | 1.79 | 1000 | 3.1729 | 1.0 |
2.9514 | 2.68 | 1500 | 2.8946 | 1.0 |
2.84 | 3.57 | 2000 | 2.8386 | 1.0 |
2.7685 | 4.46 | 2500 | 2.7147 | 1.0 |
2.5059 | 5.36 | 3000 | 2.1341 | 1.1752 |
1.8907 | 6.25 | 3500 | 1.3604 | 1.2403 |
1.3892 | 7.14 | 4000 | 0.8814 | 1.1989 |
1.0754 | 8.04 | 4500 | 0.6416 | 1.0529 |
0.8795 | 8.93 | 5000 | 0.5760 | 0.9641 |
0.7478 | 9.82 | 5500 | 0.4633 | 0.8790 |
0.6107 | 10.71 | 6000 | 0.3921 | 0.8394 |
0.5445 | 11.61 | 6500 | 0.3579 | 0.7987 |
0.4788 | 12.5 | 7000 | 0.3034 | 0.7470 |
0.4435 | 13.39 | 7500 | 0.2989 | 0.7311 |
0.4057 | 14.29 | 8000 | 0.3366 | 0.7092 |
0.3606 | 15.18 | 8500 | 0.2783 | 0.6892 |
0.343 | 16.07 | 9000 | 0.2593 | 0.6612 |
0.3189 | 16.96 | 9500 | 0.2780 | 0.6460 |
0.277 | 17.86 | 10000 | 0.3266 | 0.6277 |
0.2789 | 18.75 | 10500 | 0.3582 | 0.6253 |
0.2552 | 19.64 | 11000 | 0.3422 | 0.6156 |
0.2416 | 20.54 | 11500 | 0.3387 | 0.6016 |
0.2187 | 21.43 | 12000 | 0.3657 | 0.5845 |
0.2317 | 22.32 | 12500 | 0.2932 | 0.5845 |
0.2091 | 23.21 | 13000 | 0.2551 | 0.5614 |
0.199 | 24.11 | 13500 | 0.3113 | 0.5474 |
0.1777 | 25.0 | 14000 | 0.2895 | 0.5572 |
0.1823 | 25.89 | 14500 | 0.3127 | 0.5456 |
0.179 | 26.79 | 15000 | 0.2945 | 0.5438 |
0.1596 | 27.68 | 15500 | 0.3052 | 0.5322 |
0.1671 | 28.57 | 16000 | 0.3119 | 0.5365 |
0.1564 | 29.46 | 16500 | 0.3203 | 0.5353 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 1.18.3
- Tokenizers 0.13.2