v0.6.0-beta28
When Minutiva puts two people under one name, you can now take them apart.
Listen to every clip of a voice, not just one
A voice in the panel gets one play button and one name. That works when the separation was right, and leaves you stuck when it wasn’t — a quiet person folded in with a talkative one, dozens of clips under a single name, and one sample that can’t tell you which is which.
The … menu next to a voice now offers Review… segments. You get every clip that voice was given, with its time, what was said, and its own play button.
Move the ones that are somebody else
Tick the clips that don’t belong and send them somewhere: to another voice in the meeting, or out into a new one you can name. The transcript updates with them.
You could split off a single clip before, one at a time, from the transcript. Nobody ever finished doing that for a whole meeting — which is rather the point.
Quiet people don’t disappear any more
Voices with under two seconds of speech were hidden, because they are usually fragments the model split off by mistake. But someone who only said “Yeah” twice looks exactly the same from the outside — and once hidden, there was no way to name them or merge them.
They’re now one click away, under Show quieter voices at the bottom of the panel.
Auto-update
Existing beta channel testers get this within 24 hours. Or trigger now: Minutiva Private → Check for Updates…