Using Sussur
Live transcription & speaker labels
On-device transcription with optional speaker labels and filler cleanup.
Updated 2026-06-10 · Verified against v0.10.25
Every session is transcribed live, on your Mac, using the speech model you downloaded during setup.
On-device by default
Transcription runs entirely on your machine — your audio never leaves it. You'll see words appear as partial text that firms up into final lines as people finish speaking.
Speaker labels
Sussur can tag who's speaking on the other side (Speaker 2, Speaker 3…). Tap a label to rename it (for example, "Sarah") and it sticks for the session.
Cleaner transcripts
Optionally, Sussur can lightly polish the transcript — trimming filler words and false starts — so the post-call record reads cleanly. This step uses your AI provider; turn it off to keep transcription fully local.
Note
Only seeing your own words? That's the Screen Recording permission — it's what lets Sussur hear the other side.