Skip to main content
miniti labels remote speakers with real names where it can. instead of staring at speaker 1 / speaker 2 for the whole meeting, you’ll see jane / alex once there’s enough signal to tell them apart.
your own mic is always labeled “you” - never renamed.

​
how it works

  • miniti starts a meeting with neutral labels (speaker 1, speaker 2, …) from deepgram diarization, exactly like before
  • once there are ~8 final transcript segments, miniti runs a separate, lightweight inference pass to guess real names
  • it refreshes every ~40 new segments (with a minimum 2-minute gap) as the conversation gives more signal
  • names are overlaid on the live transcript and the speaker legend - the underlying diarization (speaker ids) doesn’t change

​
enable / disable

automatic speaker naming is a toggle under settings → notifications (alongside meeting reminders and real-time nudges), on both macos and ios. turn it off if you’d rather keep neutral speaker n labels. manual editing is always available regardless of the toggle.

​
what signals it uses

  • the transcript itself - self-introductions (“hey, this is jane”), second-person references (“jane, what do you think?”), and similar cues
  • calendar attendees - if you’ve connected google calendar, attendee names are passed as a hint list (“known attendees”). the model is told to prefer these when the conversation matches, not to invent names from them
if nobody says a name and there’s no calendar context, speakers stay as speaker 1, speaker 2, …

​
where names show up

  • live transcript - in place of speaker n once resolved
  • speaker legend / metrics - the same resolved name
  • ai insights - summary, action items, meddpicc, questions all see the resolved names
  • webhooks and markdown exports - speaker_names map is included and transcript entries carry the resolved label (see webhooks)

​
editing a name

if the model gets a name wrong (or misses one), rename the speaker yourself:
  • macos - click the speaker chip in the speaker legend above the transcript. a popover appears with the name field, plus save / clear / cancel.
  • ios - tap the speaker chip in the speaker legend. the same rename view appears as a sheet.
clear reverts to the automatic name (or back to speaker n if there is no inference yet). save pins your choice. your edit wins: once you’ve set a name, miniti will not silently rewrite it with a later automatic inference. the rename applies everywhere - live transcript, speaker legend, metrics, insights, and any subsequent webhook or markdown export. you can rename during a live meeting or on a saved meeting.

​
privacy

the transcript is sent to miniti’s insights endpoint - the same path the other modes use (standard, meddpicc, training, questions). no extra services, no extra destinations. in byok mode, the request goes directly to your openai account. the model used is gpt-5 mini. see ai insights for the full data-flow picture.

​
limits & caveats

  • names are a best guess. the model can get it wrong - especially in short meetings or when multiple attendees share a first name
  • only the remote side is named. your mic stays as “you”
  • on ios, there is no multi-speaker diarization (mic-only capture), so there’s nothing to name
  • calendar attendees are only hints. miniti will not assign a calendar attendee who never speaks