<!-- 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. -->
xls-r-300m-toi-st
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7955
- Wer: 0.8769
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: 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: 500
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.4962 | 0.3 | 500 | 0.1933 | 0.3526 |
0.4909 | 0.59 | 1000 | 0.2008 | 0.3218 |
0.4831 | 0.89 | 1500 | 0.1833 | 0.2999 |
0.4362 | 1.18 | 2000 | 0.1789 | 0.2960 |
0.423 | 1.48 | 2500 | 0.1879 | 0.3006 |
0.4129 | 1.77 | 3000 | 0.1700 | 0.2659 |
0.4567 | 2.07 | 3500 | 0.1661 | 0.2773 |
0.3837 | 2.36 | 4000 | 0.1816 | 0.2601 |
0.5766 | 2.66 | 4500 | 0.5054 | 0.6069 |
1.055 | 2.95 | 5000 | 0.7955 | 0.8769 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.10.0
- Tokenizers 0.13.2