Downloads · 30 days
2
11% of all-time downloads
Whisper-Small-Deep-Sparse-squeezeformer
jun-han/Whisper-Small-Deep-Sparse-squeezeformer
Whisper-Small-Deep-Sparse-squeezeformer is a automatic speech recognition model from jun-han. 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
19
Public
Parameters
323M
24.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.3 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Sep 25, 2024
- Updated
- Oct 3, 2024
- SHA
- 7403884c
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-relposition-Deep-Sparse-squeezeformer
This model is a fine-tuned version of openai/whisper-small on the Voice_Data_Collection_second_edition dataset. It achieves the following results on the evaluation set:
- Loss: 0.4170
- Cer: 24.9259
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 3750
- training_steps: 52500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Cer | Validation Loss |
|---|---|---|---|---|
| 2.9397 | 1.0 | 3750 | 98.2278 | 2.9583 |
| 1.4576 | 2.0 | 7500 | 77.2506 | 1.4401 |
| 0.6726 | 3.0 | 11250 | 41.7028 | 0.7099 |
| 0.4456 | 4.0 | 15000 | 33.5128 | 0.5605 |
| 0.2892 | 5.0 | 18750 | 29.2529 | 0.4902 |
| 0.1983 | 6.0 | 22500 | 27.8926 | 0.4687 |
| 0.1061 | 7.0 | 26250 | 27.1006 | 0.4629 |
| 0.0686 | 8.0 | 30000 | 26.8677 | 0.4617 |
| 0.4473 | 9.0 | 33750 | 24.1060 | 0.4217 |
| 0.3393 | 10.0 | 37500 | 23.1072 | 0.3924 |
| 0.2278 | 11.0 | 41250 | 21.9071 | 0.3780 |
| 0.3404 | 12.0 | 45000 | 23.9830 | 0.4051 |
| 0.2275 | 13.0 | 48750 | 0.4223 | 25.3396 |
| 0.2079 | 14.0 | 52500 | 0.4170 | 24.9259 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Languages
- zh
- Created
- Sep 25, 2024
- Updated
- Oct 3, 2024