Downloads · 30 days
6
18% of all-time downloads
whisper-medium-he-teamim-silsuless-ori-TrainAndVal
cantillation/whisper-medium-he-teamim-silsuless-ori-TrainAndVal
whisper-medium-he-teamim-silsuless-ori-TrainAndVal is a automatic speech recognition model from cantillation. 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
33
Public
Parameters
764M
3.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors3.1 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Feb 5, 2024
- Updated
- Feb 5, 2024
- SHA
- 58d52c11
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. -->
he
This model is a fine-tuned version of openai/whisper-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0459
- Wer: 6.3133
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: 4e-05
- 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
- training_steps: 1500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.3741 | 0.02 | 50 | 0.1764 | 6.9369 |
| 0.1107 | 0.03 | 100 | 0.1279 | 9.7428 |
| 0.1096 | 0.05 | 150 | 0.1029 | 13.0164 |
| 0.0857 | 0.07 | 200 | 0.1132 | 11.8472 |
| 0.0899 | 0.08 | 250 | 0.0909 | 6.8589 |
| 0.0814 | 0.1 | 300 | 0.1001 | 6.9369 |
| 0.0652 | 0.11 | 350 | 0.0940 | 7.0148 |
| 0.0623 | 0.13 | 400 | 0.0956 | 9.4310 |
| 0.057 | 0.15 | 450 | 0.0896 | 12.5097 |
| 0.0581 | 0.16 | 500 | 0.0841 | 5.6508 |
| 0.0518 | 0.18 | 550 | 0.0842 | 6.3913 |
| 0.0453 | 0.2 | 600 | 0.0818 | 4.8714 |
| 0.0445 | 0.21 | 650 | 0.0820 | 7.4045 |
| 0.0439 | 0.23 | 700 | 0.0741 | 9.7038 |
| 0.0456 | 0.25 | 750 | 0.0702 | 22.1356 |
| 0.032 | 0.26 | 800 | 0.0703 | 8.1450 |
| 0.0362 | 0.28 | 850 | 0.0657 | 6.3133 |
| 0.0339 | 0.29 | 900 | 0.0613 | 6.7810 |
| 0.029 | 0.31 | 950 | 0.0617 | 7.3655 |
| 0.0285 | 0.33 | 1000 | 0.0558 | 6.6641 |
| 0.0207 | 0.34 | 1050 | 0.0570 | 6.3913 |
| 0.0286 | 0.36 | 1100 | 0.0557 | 4.7545 |
| 0.0241 | 0.38 | 1150 | 0.0539 | 5.9626 |
| 0.0198 | 0.39 | 1200 | 0.0498 | 5.4170 |
| 0.0189 | 0.41 | 1250 | 0.0480 | 3.3125 |
| 0.0162 | 0.43 | 1300 | 0.0483 | 5.4949 |
| 0.0111 | 0.44 | 1350 | 0.0484 | 5.7677 |
| 0.0201 | 0.46 | 1400 | 0.0465 | 5.6118 |
| 0.0124 | 0.48 | 1450 | 0.0462 | 6.2744 |
| 0.0214 | 0.49 | 1500 | 0.0459 | 6.3133 |
Framework versions
- Transformers 4.37.0.dev0
- Pytorch 2.1.2
- Datasets 2.16.1
- Tokenizers 0.15.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Languages
- he
- Created
- Feb 5, 2024
- Updated
- Feb 5, 2024