Fact · Speech recognition
4times faster, at most
faster-whisper is up to 4 times faster than openai/whisper for the same accuracy, while using less memory.
Checked against its source on October 17, 2026
In the source's own words
“This implementation is up to 4 times faster than ... for the same accuracy while using less memory.”
faster-whisper: README (GitHub, SYSTRAN), SYSTRAN, GitHub (SYSTRAN/faster-whisper), 2026, Opening paragraph.
Open the original- What it measures
- faster-whisper reimplements Whisper on the CTranslate2 inference engine. The project's README gives this claim before its benchmark tables.
- What it does not show
- A figure from the project's own README. The gain depends on hardware, precision and beam size, and the benchmark tables show it on one 13-minute recording.