Downloads · 30 days
9
30% of all-time downloads
whisper-tiny-vi
hieulhwork24/whisper-tiny-vi
whisper-tiny-vi is a automatic speech recognition model from hieulhwork24. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
30
Public
Parameters
37.8M
1.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors151 MB · 97%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Mar 18, 2025
- Updated
- Mar 18, 2025
- SHA
- 649622a6
Base models
Datasets
Model card
From the Hugging Face model README
<!-- 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. -->
whisper-tiny-vi
This model is a fine-tuned version of openai/whisper-tiny on the vivos dataset. It achieves the following results on the evaluation set:
- Loss: 0.1308
- Wer: 26.0096
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 4.2251 | 0.8576 | 500 | 0.3272 | 57.3497 |
| 0.1917 | 1.7153 | 1000 | 0.1825 | 38.6898 |
| 0.1244 | 2.5729 | 1500 | 0.1537 | 30.6389 |
| 0.0954 | 3.4305 | 2000 | 0.1428 | 28.8859 |
| 0.0758 | 4.2882 | 2500 | 0.1362 | 24.8961 |
| 0.0642 | 5.1458 | 3000 | 0.1331 | 28.9378 |
| 0.0537 | 6.0034 | 3500 | 0.1308 | 26.0096 |
| 0.0446 | 6.8611 | 4000 | 0.1325 | 24.8766 |
| 0.0406 | 7.7187 | 4500 | 0.1329 | 25.0 |
| 0.0365 | 8.5763 | 5000 | 0.1331 | 25.7207 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.6.0+cu124
- Datasets 3.4.1
- Tokenizers 0.21.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Mar 18, 2025
- Updated
- Mar 18, 2025