Fact · Speech recognition
59seconds
On the same GPU, faster-whisper with int8 precision transcribed the recording with the large-v2 model in 59 seconds.
Checked against its source on October 17, 2026
In the source's own words
“| faster-whisper | int8 | 5 | 59s | 2926MB |”
faster-whisper: README (GitHub, SYSTRAN), SYSTRAN, GitHub (SYSTRAN/faster-whisper), 2026, Large-v2 model on GPU table.
Open the original- What it measures
- The same table lists faster-whisper in fp16 at 1m03s with 4525 MB of VRAM, and int8 with 2926 MB of VRAM, which is the lowest memory figure in the table.
- What it does not show
- A single run on one GPU. Int8 is a lower-precision mode, and the README reports the time, not a separate accuracy score.