Downloads · 30 days
3
4% of all-time downloads
ap-gQPvCZE1ClYjiHJYVrmkTQ
dhaval0108/ap-gQPvCZE1ClYjiHJYVrmkTQ
ap-gQPvCZE1ClYjiHJYVrmkTQ is a automatic speech recognition model from dhaval0108. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
All-time downloads
80
Public
Parameters
809M
3.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.2 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- mit
- Model type
- whisper
- Access
- Public
- Created
- Jul 30, 2025
- Updated
- Jul 30, 2025
- SHA
- e22bbe25
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. -->
<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/> <img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>
ap-gQPvCZE1ClYjiHJYVrmkTQ
This model is a fine-tuned version of openai/whisper-large-v3-turbo on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5599
- Model Preparation Time: 0.0074
- Wer: 0.1343
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: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- 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: 400
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Wer |
|---|---|---|---|---|---|
| 0.7748 | 1.0 | 96 | 0.3779 | 0.0074 | 0.1654 |
| 0.2836 | 2.0 | 192 | 0.4260 | 0.0074 | 0.1973 |
| 0.2354 | 3.0 | 288 | 0.3930 | 0.0074 | 0.1524 |
| 0.2097 | 4.0 | 384 | 0.4422 | 0.0074 | 0.2027 |
| 0.1942 | 5.0 | 480 | 0.4359 | 0.0074 | 0.2456 |
| 0.1625 | 6.0 | 576 | 0.4169 | 0.0074 | 0.1910 |
| 0.1318 | 7.0 | 672 | 0.4465 | 0.0074 | 0.1603 |
| 0.1106 | 8.0 | 768 | 0.4704 | 0.0074 | 0.1860 |
| 0.0946 | 9.0 | 864 | 0.4792 | 0.0074 | 0.1685 |
| 0.0801 | 10.0 | 960 | 0.4609 | 0.0074 | 0.1617 |
| 0.0711 | 11.0 | 1056 | 0.4906 | 0.0074 | 0.1621 |
| 0.0569 | 12.0 | 1152 | 0.5165 | 0.0074 | 0.1690 |
| 0.0471 | 13.0 | 1248 | 0.5087 | 0.0074 | 0.1492 |
| 0.0374 | 14.0 | 1344 | 0.5130 | 0.0074 | 0.1348 |
| 0.0337 | 15.0 | 1440 | 0.5367 | 0.0074 | 0.1394 |
| 0.0256 | 16.0 | 1536 | 0.5285 | 0.0074 | 0.1350 |
| 0.0241 | 17.0 | 1632 | 0.5178 | 0.0074 | 0.1370 |
| 0.0143 | 18.0 | 1728 | 0.5489 | 0.0074 | 0.1283 |
| 0.0114 | 19.0 | 1824 | 0.5605 | 0.0074 | 0.1349 |
| 0.0066 | 19.7927 | 1900 | 0.5599 | 0.0074 | 0.1343 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.2
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- mit
- Created
- Jul 30, 2025
- Updated
- Jul 30, 2025