<!-- 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. -->
v2-fine-tune-wav2vec2-Vietnamese-ARS-demo
This model is a fine-tuned version of nguyenvulebinh/wav2vec2-base-vietnamese-250h on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2515
- Wer: 0.2235
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: 6
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
8.8651 | 0.34 | 500 | 3.6919 | 0.9999 |
3.54 | 0.69 | 1000 | 3.3584 | 1.0 |
2.9478 | 1.03 | 1500 | 2.2535 | 0.9885 |
1.9147 | 1.37 | 2000 | 0.9977 | 0.7260 |
1.1667 | 1.71 | 2500 | 0.5577 | 0.4746 |
0.844 | 2.06 | 3000 | 0.4129 | 0.3581 |
0.6968 | 2.4 | 3500 | 0.3566 | 0.3090 |
0.6273 | 2.74 | 4000 | 0.3243 | 0.2813 |
0.5434 | 3.09 | 4500 | 0.3076 | 0.2631 |
0.5069 | 3.43 | 5000 | 0.2902 | 0.2539 |
0.4842 | 3.77 | 5500 | 0.2752 | 0.2432 |
0.4318 | 4.12 | 6000 | 0.2854 | 0.2384 |
0.3951 | 4.46 | 6500 | 0.2674 | 0.2350 |
0.3954 | 4.8 | 7000 | 0.2628 | 0.2322 |
0.3763 | 5.14 | 7500 | 0.2609 | 0.2284 |
0.3652 | 5.49 | 8000 | 0.2508 | 0.2249 |
0.3703 | 5.83 | 8500 | 0.2515 | 0.2235 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.12.0+cu113
- Datasets 1.18.3
- Tokenizers 0.12.1