Downloads · 30 days
4
9% of all-time downloads
Hanhpt23/whisper-small-germanmed-free_E0-8
whisper-small-germanmed-free_E0-8 is a automatic speech recognition model from Hanhpt23. 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. --
Downloads · 30 days
4
9% of all-time downloads
All-time downloads
47
Public
Parameters
242M
35.8 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors967 MB · 100%
From the Hugging Face model README
This model is a fine-tuned version of openai/whisper-small on the Hanhpt23/GermanMed-full dataset. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.6088 | 1.0 | 194 | 0.5708 | 42.7234 | 27.2387 |
| 0.2668 | 2.0 | 388 | 0.5730 | 30.4536 | 18.2579 |
| 0.1536 | 3.0 | 582 | 0.5971 | 30.4330 | 19.1535 |
| 0.1118 | 4.0 | 776 | 0.6131 | 25.2700 | 15.6419 |
| 0.0878 | 5.0 | 970 | 0.6541 | 28.3657 | 17.5805 |
| 0.0574 | 6.0 | 1164 | 0.6766 | 25.7431 | 16.3089 |
| 0.0417 | 7.0 | 1358 | 0.6932 | 29.3737 | 18.7464 |
| 0.0426 | 8.0 | 1552 | 0.6984 | 24.9820 | 16.0750 |
| 0.0225 | 9.0 | 1746 | 0.6977 | 25.9282 | 16.3107 |
| 0.0199 | 10.0 | 1940 | 0.7123 | 24.4678 | 15.1725 |
| 0.0134 | 11.0 | 2134 | 0.7021 | 24.6220 | 15.4254 |
| 0.0089 | 12.0 | 2328 | 0.6977 | 24.7557 | 15.4289 |
| 0.0057 | 13.0 | 2522 | 0.6929 | 24.0975 | 14.9784 |
| 0.0048 | 14.0 | 2716 | 0.7050 | 24.1078 | 15.4462 |
| 0.0009 | 15.0 | 2910 | 0.7049 | 24.1078 | 15.4964 |
| 0.0002 | 16.0 | 3104 | 0.7080 | 23.6553 | 15.0391 |
| 0.0002 | 17.0 | 3298 | 0.7083 | 25.3625 | 15.4063 |
| 0.0001 | 18.0 | 3492 | 0.7106 | 25.3214 | 15.3630 |
| 0.0002 | 19.0 | 3686 | 0.7114 | 25.3831 | 15.3752 |
| 0.0003 | 20.0 | 3880 | 0.7117 | 25.4037 | 15.3838 |