<!-- 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-demo-colab
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5752
- Wer: 0.8492
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: 0.0003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 30
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
5.8188 | 1.82 | 400 | 3.3642 | 1.0 |
1.4611 | 3.64 | 800 | 0.4218 | 0.4522 |
0.5377 | 5.47 | 1200 | 0.3572 | 0.3881 |
0.4306 | 7.29 | 1600 | 0.2934 | 0.3306 |
0.3689 | 9.11 | 2000 | 0.2787 | 0.2980 |
0.325 | 10.93 | 2400 | 0.2893 | 0.3141 |
0.2956 | 12.76 | 2800 | 0.2615 | 0.2753 |
0.2781 | 14.58 | 3200 | 0.2859 | 0.2822 |
0.3729 | 16.4 | 3600 | 0.4245 | 0.3831 |
0.5928 | 18.22 | 4000 | 0.4798 | 0.4108 |
0.7275 | 20.05 | 4400 | 0.8048 | 0.6285 |
1.1241 | 21.87 | 4800 | 0.8472 | 0.6749 |
1.4631 | 23.69 | 5200 | 1.3780 | 0.8773 |
1.5766 | 25.51 | 5600 | 1.3799 | 0.8594 |
1.675 | 27.33 | 6000 | 1.4853 | 0.8546 |
1.7743 | 29.16 | 6400 | 1.5752 | 0.8492 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3