metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: wav2vec_asr_swbd
results: []
wav2vec_asr_swbd
This model is a fine-tuned version of facebook/wav2vec2-large-robust-ft-swbd-300h on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5060
- Wer: 0.3620
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: 7
- 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: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.8809 | 0.44 | 10000 | 0.6088 | 0.4008 |
| 0.8435 | 0.77 | 20000 | 0.6319 | 0.3914 |
| 0.7454 | 1.15 | 30000 | 0.6126 | 0.3809 |
| 0.6806 | 1.53 | 40000 | 0.5307 | 0.3792 |
| 0.6589 | 1.91 | 50000 | 0.5430 | 0.3710 |
| 0.603 | 2.3 | 60000 | 0.5298 | 0.3646 |
| 0.5886 | 2.68 | 70000 | 0.5060 | 0.3620 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.11.0+cu113
- Datasets 1.18.4
- Tokenizers 0.11.6