Automatically transcribe speech to text in your audio or video files with Jojo Transcribe!
Powered by the popular Whisper model and Whisper.cpp.
- Automatically transcribes and translates 100+ languages
- Line-by-line playback and audio export
- Quickly and easily make manual corrections where necessary
- Export text with configurable timestamp frequency
- Export subtitles in srt and vtt format
- Save and load Jojo documents and play back previously processed files, synced to the text
- Download additional models, such as NB-Whisper by Nasjonalbiblioteket
- Everything happens locally on your machine. Once downloaded, no internet connection is required.
Jojo was developed as a tool for journalists working in the Norwegian newspaper VG and is by default using the largest version of Whisper.
This gives the most accurate results for the Norwegian (and other languages not represented with huge amounts of training data).
It also means transcriptions may take a while and use a lot of processing power on your machine.