|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: XLS-R_timit_en |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- 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_timit_en |
|
|
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.2960 |
|
|
- Wer: 0.2705 |
|
|
|
|
|
## 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: 8 |
|
|
- 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: 30 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | |
|
|
|:-------------:|:-----:|:-----:|:---------------:|:------:| |
|
|
| 3.6358 | 2.01 | 1000 | 0.7983 | 0.7896 | |
|
|
| 0.6096 | 4.02 | 2000 | 0.2907 | 0.3794 | |
|
|
| 0.314 | 6.02 | 3000 | 0.2625 | 0.3246 | |
|
|
| 0.2259 | 8.03 | 4000 | 0.2673 | 0.3058 | |
|
|
| 0.1771 | 10.04 | 5000 | 0.2518 | 0.2932 | |
|
|
| 0.1474 | 12.05 | 6000 | 0.2717 | 0.2900 | |
|
|
| 0.1267 | 14.06 | 7000 | 0.2700 | 0.2821 | |
|
|
| 0.1069 | 16.06 | 8000 | 0.2941 | 0.2834 | |
|
|
| 0.0991 | 18.07 | 9000 | 0.3021 | 0.2806 | |
|
|
| 0.0853 | 20.08 | 10000 | 0.3088 | 0.2803 | |
|
|
| 0.0787 | 22.09 | 11000 | 0.2987 | 0.2770 | |
|
|
| 0.067 | 24.1 | 12000 | 0.3182 | 0.2734 | |
|
|
| 0.0652 | 26.1 | 13000 | 0.3117 | 0.2701 | |
|
|
| 0.0636 | 28.11 | 14000 | 0.2960 | 0.2705 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.17.0 |
|
|
- Pytorch 1.12.1+cu113 |
|
|
- Datasets 1.18.3 |
|
|
- Tokenizers 0.12.1 |
|
|
|
