metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-large-asr-th
results: []
wav2vec2-large-asr-th
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5207
- Wer: 0.4689
- Cer: 0.1571
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 3500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 3.6231 | 0.57 | 500 | 3.5254 | 1.0 | 0.9999 |
| 1.5065 | 1.14 | 1000 | 1.0542 | 0.7677 | 0.2939 |
| 1.0592 | 1.71 | 1500 | 0.7072 | 0.5837 | 0.2054 |
| 1.0818 | 2.28 | 2000 | 0.6309 | 0.5414 | 0.1887 |
| 1.0136 | 2.85 | 2500 | 0.5573 | 0.4966 | 0.1690 |
| 0.7907 | 3.42 | 3000 | 0.5207 | 0.4689 | 0.1571 |
| 0.8223 | 4.0 | 3500 | 0.5079 | 0.4642 | 0.1546 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.9.0
- Tokenizers 0.13.2