Paste a labeled transcript
Names on the lines, as in a Zoom VTT with speakers.
Runs in this browser · Nothing uploaded · No model run
Who spoke the most?
From a labeled transcript.
Paste a transcript that already has speaker names and see each name’s share of the words. If there are no names, the page does not guess.
Runs in this browser. The paste never leaves this device.
Speaker talk time from a transcript
This is not a score. It is a word count per name already written in the file. Facilitation metrics stay with you. A fiduciary or chair may want the split. A raw M4A cannot provide it here.
Who spoke the most, with no names in the file
The page stops. It will not invent Speaker 1. Diarization is a Mac job. Clean a VTT first only if names are already on the cues.
Speaking time without the timestamps
Timestamps in a VTT are for a player. Cue duration is not speaking time: silence and overlap sit on the clock too. Without clocks the page would have to invent them. Words on labeled lines are the honest count.
What the Mac app adds
Vemoir separates speakers on device from the recording, then you can rename them. This page only reads labels that already exist.
Who is speaking?
This page never infers speakers. No labels means an honest stop, not a fake Speaker 1.
The count does not restore dialect
Counts follow the text you paste. Dialect is not restored.
How to see who spoke the most
Show who held the floor
Each name gets a share of the words.
If there are no names, stop
The page will not guess. Use Vemoir for Mac on the recording.
Before you start
Does this separate speakers from audio?
No. Paste text that already has names.
Does it spend a weekly run?
No. There is no speech model on this page.
What if only some lines have names?
Counts use the labeled lines. Unlabeled lines are ignored rather than assigned.
Is a high share a problem?
The page does not judge. It reports the split.
Can I use a Zoom VTT?
Yes, if the VTT already contains names. Cue clocks are ignored. Clean timestamps first if you want readable text as well.
Is the paste uploaded?
No.
Does this work in Swiss German?
It counts whatever text you paste. It does not restore dialect.
Where do unlabeled recordings go?
Vemoir for Mac, for speaker separation on the machine.
Get names from the recording itself.
Vemoir for Mac separates speakers on the machine. This page only reads labels you already have.