Downloads · 30 days
2
5% of all-time downloads
whisper-small-E30_pause
Gummybear05/whisper-small-E30_pause
whisper-small-E30_pause is a automatic speech recognition model from Gummybear05. 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
38
Public
Parameters
242M
15.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors967 MB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Oct 11, 2024
- Updated
- Oct 12, 2024
- SHA
- cc80792c
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-small-Yfreq_speed_pause
This model is a fine-tuned version of openai/whisper-small on the aihub old30 adult pause changed dataset. It achieves the following results on the evaluation set:
- Loss: 0.1831
- Cer: 5.5040
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- 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: 50
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.2738 | 0.1289 | 100 | 0.2498 | 6.3675 |
| 0.1412 | 0.2579 | 200 | 0.2281 | 6.2089 |
| 0.1672 | 0.3868 | 300 | 0.2250 | 6.0444 |
| 0.1311 | 0.5158 | 400 | 0.2185 | 6.0092 |
| 0.1183 | 0.6447 | 500 | 0.2125 | 5.7213 |
| 0.1053 | 0.7737 | 600 | 0.1996 | 5.2338 |
| 0.1035 | 0.9026 | 700 | 0.1952 | 5.3160 |
| 0.0566 | 1.0316 | 800 | 0.1894 | 5.1574 |
| 0.0476 | 1.1605 | 900 | 0.1922 | 5.0869 |
| 0.0373 | 1.2895 | 1000 | 0.1887 | 5.1868 |
| 0.0379 | 1.4184 | 1100 | 0.1890 | 5.3924 |
| 0.0404 | 1.5474 | 1200 | 0.1870 | 6.0444 |
| 0.038 | 1.6763 | 1300 | 0.1842 | 5.6039 |
| 0.0411 | 1.8053 | 1400 | 0.1846 | 5.5275 |
| 0.0366 | 1.9342 | 1500 | 0.1831 | 5.5040 |
Framework versions
- Transformers 4.46.0.dev0
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Languages
- hi
- Created
- Oct 11, 2024
- Updated
- Oct 12, 2024