|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- wer |
|
model-index: |
|
- name: wav2vec2-large-asr-th |
|
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. --> |
|
|
|
# wav2vec2-large-asr-th |
|
|
|
This model was trained from scratch on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4214 |
|
- Wer: 0.3708 |
|
- Cer: 0.1236 |
|
|
|
## 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: 0.0002 |
|
- train_batch_size: 24 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 48 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 100 |
|
- training_steps: 3000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:| |
|
| 0.6395 | 0.86 | 500 | 0.5238 | 0.4279 | 0.1463 | |
|
| 0.4942 | 1.71 | 1000 | 0.5227 | 0.4188 | 0.1404 | |
|
| 0.4195 | 2.57 | 1500 | 0.4984 | 0.4019 | 0.1344 | |
|
| 0.514 | 3.42 | 2000 | 0.4713 | 0.3828 | 0.1305 | |
|
| 0.4964 | 4.28 | 2500 | 0.4490 | 0.3780 | 0.1261 | |
|
| 0.5175 | 5.14 | 3000 | 0.4214 | 0.3708 | 0.1236 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1+cu116 |
|
- Datasets 2.10.1 |
|
- Tokenizers 0.13.2 |
|
|
