<!-- 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. -->
torgo_xlsr_finetune-M02-2
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: 2.1942
- Wer: 1.0791
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 |
|---|---|---|---|---|
| 23.2194 | 0.91 | 500 | 3.2975 | 0.9820 |
| 3.3856 | 1.81 | 1000 | 3.2225 | 0.9820 |
| 2.9403 | 2.72 | 1500 | 2.7805 | 0.9820 |
| 2.6255 | 3.62 | 2000 | 2.3830 | 0.9834 |
| 1.8901 | 4.53 | 2500 | 1.7920 | 1.3779 |
| 1.2594 | 5.43 | 3000 | 1.7845 | 1.3343 |
| 1.0008 | 6.34 | 3500 | 1.6923 | 1.3148 |
| 0.7896 | 7.25 | 4000 | 1.5444 | 1.2822 |
| 0.6373 | 8.15 | 4500 | 1.5547 | 1.2670 |
| 0.5639 | 9.06 | 5000 | 1.5924 | 1.1935 |
| 0.5 | 9.96 | 5500 | 1.7545 | 1.2060 |
| 0.4488 | 10.87 | 6000 | 1.6170 | 1.1498 |
| 0.3892 | 11.78 | 6500 | 1.7550 | 1.1664 |
| 0.3497 | 12.68 | 7000 | 1.9707 | 1.1845 |
| 0.3444 | 13.59 | 7500 | 1.8976 | 1.1657 |
| 0.3153 | 14.49 | 8000 | 1.9255 | 1.1415 |
| 0.2967 | 15.4 | 8500 | 1.8710 | 1.1470 |
| 0.2698 | 16.3 | 9000 | 1.7480 | 1.1408 |
| 0.2661 | 17.21 | 9500 | 1.8134 | 1.1123 |
| 0.255 | 18.12 | 10000 | 2.0806 | 1.1248 |
| 0.2021 | 19.02 | 10500 | 2.1426 | 1.1172 |
| 0.2179 | 19.93 | 11000 | 2.0832 | 1.1117 |
| 0.193 | 20.83 | 11500 | 2.3892 | 1.1262 |
| 0.1999 | 21.74 | 12000 | 2.1183 | 1.1096 |
| 0.1857 | 22.64 | 12500 | 1.9986 | 1.1047 |
| 0.1804 | 23.55 | 13000 | 2.0788 | 1.0978 |
| 0.1568 | 24.46 | 13500 | 2.2646 | 1.1026 |
| 0.1596 | 25.36 | 14000 | 2.0863 | 1.0673 |
| 0.1553 | 26.27 | 14500 | 2.1758 | 1.0804 |
| 0.1484 | 27.17 | 15000 | 2.0750 | 1.0693 |
| 0.1359 | 28.08 | 15500 | 2.1412 | 1.0784 |
| 0.131 | 28.99 | 16000 | 2.2616 | 1.0853 |
| 0.1386 | 29.89 | 16500 | 2.1942 | 1.0791 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 1.18.3
- Tokenizers 0.13.2
