metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: wav2vec2-base-timit-eng
results: []
wav2vec2-base-timit-eng
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: 1.2054
- Wer: 0.5205
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: 16
- 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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.5248 | 2.01 | 500 | 2.2003 | 0.9990 |
| 0.7619 | 4.02 | 1000 | 1.4549 | 0.6285 |
| 0.3232 | 6.02 | 1500 | 1.2368 | 0.5479 |
| 0.1934 | 8.03 | 2000 | 1.2054 | 0.5205 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.13.0+cu116
- Datasets 1.18.3
- Tokenizers 0.13.2