Downloads · 30 days
7
3% of all-time downloads
whisper-large-fo-100h-30k-steps
davidilag/whisper-large-fo-100h-30k-steps
whisper-large-fo-100h-30k-steps is a automatic speech recognition model from davidilag. 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
206
Public
Parameters
1.5B
6.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors6.2 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Jul 19, 2024
- Updated
- Jul 21, 2024
- SHA
- 3ec5503d
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-large-fo-100h-30k-steps
This model is a fine-tuned version of openai/whisper-large on the ravnursson_asr dataset. It achieves the following results on the evaluation set:
- Loss: 0.0872
- Wer: 4.9577
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: 500
- training_steps: 30000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2261 | 0.2320 | 1000 | 0.2668 | 20.1379 |
| 0.1577 | 0.4640 | 2000 | 0.1840 | 15.0997 |
| 0.1205 | 0.6961 | 3000 | 0.1456 | 11.9489 |
| 0.1151 | 0.9281 | 4000 | 0.1300 | 10.6906 |
| 0.0457 | 1.1601 | 5000 | 0.1241 | 9.7745 |
| 0.0423 | 1.3921 | 6000 | 0.1221 | 9.4876 |
| 0.0428 | 1.6241 | 7000 | 0.1080 | 8.4709 |
| 0.0486 | 1.8561 | 8000 | 0.1053 | 8.5011 |
| 0.0205 | 2.0882 | 9000 | 0.1014 | 7.4643 |
| 0.0184 | 2.3202 | 10000 | 0.1003 | 8.1387 |
| 0.0165 | 2.5522 | 11000 | 0.0969 | 7.1472 |
| 0.025 | 2.7842 | 12000 | 0.0907 | 6.8804 |
| 0.0048 | 3.0162 | 13000 | 0.0936 | 6.9005 |
| 0.0092 | 3.2483 | 14000 | 0.0923 | 6.7244 |
| 0.006 | 3.4803 | 15000 | 0.0921 | 6.3519 |
| 0.0095 | 3.7123 | 16000 | 0.0922 | 6.3821 |
| 0.0089 | 3.9443 | 17000 | 0.0929 | 6.3771 |
| 0.0023 | 4.1763 | 18000 | 0.0915 | 6.0650 |
| 0.0033 | 4.4084 | 19000 | 0.0924 | 5.9543 |
| 0.0028 | 4.6404 | 20000 | 0.0909 | 5.9040 |
| 0.0021 | 4.8724 | 21000 | 0.0884 | 5.7328 |
| 0.002 | 5.1044 | 22000 | 0.0874 | 5.4057 |
| 0.0008 | 5.3364 | 23000 | 0.0890 | 5.3654 |
| 0.0005 | 5.5684 | 24000 | 0.0857 | 5.2597 |
| 0.002 | 5.8005 | 25000 | 0.0860 | 5.2144 |
| 0.0007 | 6.0325 | 26000 | 0.0873 | 5.1842 |
| 0.0002 | 6.2645 | 27000 | 0.0850 | 4.9879 |
| 0.001 | 6.4965 | 28000 | 0.0889 | 4.9376 |
| 0.0001 | 6.7285 | 29000 | 0.0878 | 5.0081 |
| 0.0003 | 6.9606 | 30000 | 0.0872 | 4.9577 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Jul 19, 2024
- Updated
- Jul 21, 2024