<!-- 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_toy_train_data_random_low_pass
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: 0.6572
- Wer: 0.4973
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
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
3.0834 | 2.1 | 500 | 3.4478 | 1.0 |
1.0735 | 4.2 | 1000 | 0.9113 | 0.7815 |
0.5516 | 6.3 | 1500 | 0.7035 | 0.6081 |
0.4023 | 8.4 | 2000 | 0.6647 | 0.5649 |
0.3423 | 10.5 | 2500 | 0.6613 | 0.5450 |
0.2938 | 12.6 | 3000 | 0.6967 | 0.5318 |
0.2902 | 14.7 | 3500 | 0.6430 | 0.5089 |
0.2372 | 16.81 | 4000 | 0.6653 | 0.5045 |
0.2148 | 18.91 | 4500 | 0.6572 | 0.4973 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.11.0+cu102
- Datasets 2.0.0
- Tokenizers 0.12.0