metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: wav2vec2-base-timit-demo-colab-1
results: []
wav2vec2-base-timit-demo-colab-1
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6895
- Wer: 0.4357
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: 6
- 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: 800
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.6622 | 5.26 | 500 | 3.1853 | 1.0 |
| 1.46 | 10.53 | 1000 | 0.7476 | 0.5737 |
| 0.497 | 15.79 | 1500 | 0.6507 | 0.4953 |
| 0.3041 | 21.05 | 2000 | 0.6530 | 0.4546 |
| 0.2073 | 26.32 | 2500 | 0.6895 | 0.4357 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.11.0+cu113
- Datasets 1.18.3
- Tokenizers 0.10.3