Speaker profiles
Abschrift doesn't just separate speakers within one meeting — it remembers voices across meetings. This page explains how that works and how to keep your speaker list tidy.
How voice memory works
During processing, each remote speaker's voice gets a compact acoustic fingerprint (an
embedding — not a recording). Abschrift compares new voices against your stored profiles: a
match fills in the name automatically; a genuinely new voice becomes a new profile named
“Speaker N”. All of this happens on your Mac; fingerprints live in
~/Abschrift/speakers.sqlite and never leave it.
Your own voice is special-cased: the microphone track is always attributed to You.
Renaming
Open Speakers from the sidebar footer and type over a name. Renames propagate into all past meetings — transcript labels, the participants list, and search — so history always reflects what you know now. (Prose inside notes is only rewritten for auto-assigned “Speaker N” names; sentences mentioning real names are left untouched.)
Merging duplicates
Occasionally one person ends up as two profiles (bad headset day, very short first appearance). Use Merge into… on the duplicate: the two voice fingerprints are combined, the duplicate disappears, and its past meetings are relabeled with the surviving name. Future meetings recognize the person more reliably after a merge — you gave Abschrift more voice evidence to work with.
Deleting
Deleting a profile removes the stored voice fingerprint. Past transcripts keep their labels; the person is simply treated as unknown next time (they'll re-appear as “Speaker N”).
Tips for best results
- Name speakers early — every named meeting strengthens recognition.
- If someone is misrecognized once, fix the profile (rename/merge) rather than editing the file: the store learns, files don't.
- The same tools exist headlessly:
abschrift speakers list|rename|merge|delete.