metadata
license: apache-2.0
tags:
- generated_from_trainer
- audio
- automatic-speech-recognition
- speech
pipeline_tag: automatic-speech-recognition
model-index:
name: iloko_model
iloko_model
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: 0.0663
- Wer: 0.1235
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.4454 | 1.11 | 100 | 3.0569 | 1.0 |
| 2.6883 | 2.22 | 200 | 2.6725 | 1.0 |
| 2.6082 | 3.33 | 300 | 2.6133 | 1.0 |
| 1.7907 | 4.44 | 400 | 1.4020 | 0.9996 |
| 0.5616 | 5.56 | 500 | 0.2521 | 0.3217 |
| 0.2554 | 6.67 | 600 | 0.1373 | 0.1981 |
| 0.1604 | 7.78 | 700 | 0.0904 | 0.1468 |
| 0.1507 | 8.89 | 800 | 0.0740 | 0.1308 |
| 0.1012 | 10.0 | 900 | 0.0663 | 0.1235 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu102
- Datasets 1.13.3
- Tokenizers 0.10.3