Downloads · 30 days
16
14% of all-time downloads
vibase-fineturn
dolphinnlp/vibase-fineturn
vibase-fineturn is a automatic speech recognition model from dolphinnlp. 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
116
Public
Parameters
72.6M
581 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors290 MB · 99%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Jan 12, 2025
- Updated
- Jan 23, 2025
- SHA
- 38c8b21d
Base models
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 small vi - Ox
This model is a fine-tuned version of openai/whisper-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3667
- Wer: 33.5754
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 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
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.7172 | 0.1028 | 1000 | 0.7280 | 40.3137 |
| 0.6097 | 0.2055 | 2000 | 0.6292 | 35.9280 |
| 0.6005 | 0.3083 | 3000 | 0.5747 | 39.8925 |
| 0.5383 | 0.4111 | 4000 | 0.5368 | 51.1182 |
| 0.5332 | 0.5138 | 5000 | 0.5145 | 49.7822 |
| 0.4759 | 0.6166 | 6000 | 0.4885 | 70.9410 |
| 0.4824 | 0.7194 | 7000 | 0.4662 | 123.4679 |
| 0.437 | 0.8221 | 8000 | 0.4551 | 77.9988 |
| 0.4217 | 0.9249 | 9000 | 0.4411 | 60.1220 |
| 0.3637 | 1.0276 | 10000 | 0.4334 | 78.2602 |
| 0.3461 | 1.1304 | 11000 | 0.4260 | 46.1226 |
| 0.3613 | 1.2332 | 12000 | 0.4175 | 47.5748 |
| 0.3478 | 1.3359 | 13000 | 0.4104 | 32.2829 |
| 0.3413 | 1.4387 | 14000 | 0.4033 | 30.0465 |
| 0.3331 | 1.5415 | 15000 | 0.4022 | 24.7023 |
| 0.3325 | 1.6442 | 16000 | 0.3935 | 24.2085 |
| 0.3142 | 1.7470 | 17000 | 0.3921 | 24.8185 |
| 0.3158 | 1.8498 | 18000 | 0.3847 | 26.9532 |
| 0.305 | 1.9525 | 19000 | 0.3817 | 26.2416 |
| 0.2887 | 2.0553 | 20000 | 0.3822 | 26.4304 |
| 0.2723 | 2.1581 | 21000 | 0.3793 | 28.8411 |
| 0.3019 | 2.2608 | 22000 | 0.3756 | 33.7206 |
| 0.2693 | 2.3636 | 23000 | 0.3757 | 32.0796 |
| 0.2721 | 2.4663 | 24000 | 0.3760 | 31.7746 |
| 0.2715 | 2.5691 | 25000 | 0.3727 | 34.9695 |
| 0.272 | 2.6719 | 26000 | 0.3716 | 34.1708 |
| 0.2557 | 2.7746 | 27000 | 0.3702 | 34.7517 |
| 0.2661 | 2.8774 | 28000 | 0.3676 | 34.0546 |
| 0.2522 | 2.9802 | 29000 | 0.3667 | 33.5754 |
Framework versions
- Transformers 4.48.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Languages
- vi
- Created
- Jan 12, 2025
- Updated
- Jan 23, 2025