Meeting transcription
Every word, every speaker, a finished report.
Drop in a recording, record the room, or capture an online call. Butler writes down who said what, names the people it knows, and turns the meeting into a report you can edit and export — without the audio ever leaving your server.
- Whisper large-v3 + speaker detection
- Report in Word or PDF
- On your own server
Three ways in
From a file, from the room, or from any online call.
Butler never joins your meeting as a bot. It records the sound your computer already plays, so it works with every meeting application.
A recording you already have
MP3, WAV, M4A, MP4 and more — or paste a link to a video or podcast and let the server fetch only the audio.
The room
Your microphone, with pause and resume. The audio is saved to the server every minute, so a crashed tab loses nothing.
An online call
Zoom, Teams, Meet, Webex: your microphone on one channel, the call on the other — so your own words are labelled with your name.
Who said what
Speakers detected, then named.
Each passage is labelled with its speaker. Teach Butler the voices of your team once, and the next transcripts carry real names instead of “Speaker 1”.
Correct in place
Click a sentence to fix it, click a name to rename that speaker everywhere.
Voice dictionaries
A few samples per person; Butler compares every new meeting against them.
Raw and AI-polished
Keep the verbatim, or let the AI fix wording and merge a speaker’s fragments.
Timed to the audio
Every passage carries its time stamps, with the recording beside the transcript.
From transcript to report
The meeting, written up the way your organisation writes.
One click turns the transcript into a report: summary, participants, topics, decisions, actions and open questions. Your own templates decide what goes in and how it reads.
Your templates
Board meeting, project review, research interview: fixed sections, tone and language, shared with your team.
Decisions and actions
Who does what, by when — pulled from what was actually said.
Word or PDF
Export the report alone or with the full transcript, exactly as you saved it.
What people do with it
Meetings that leave a record.
Project and steering meetings
Decisions, owners and dates written down the same day, in the format the team already uses.
Research interviews
Hours of interviews transcribed with speakers, then searched and quoted in a knowledge base.
Councils and committees
Minutes drafted from the recording, with every speaker identified.
Lectures and talks
A transcript and a summary for those who missed it, in the language of the talk.
Your recordings stay yours
Transcribed where the audio is.
Local speech recognition
Whisper large-v3 and speaker detection run on Butler’s own GPUs — no transcription API, no cloud.
Consent first
Recording a call starts only once you confirm that everyone has been told.
Nothing used for training
Not one recording trains a model — not ours, not anyone else’s.
Try it on one of your meetings.
Send us a recording you are allowed to share — we show you the transcript and the report Butler makes of it.
Questions
Which languages?
Every language Whisper large-v3 knows, including English, German and French. Setting the language and the number of speakers gives the best first transcript.
Does a bot join our calls?
No. Butler records the sound this computer already plays: it works with every meeting application, and none of them can block it.
How long does it take?
The two-minute meeting in the recordings took under twenty seconds once the models were loaded. It runs in the background, so you can close the page.
Can we share meetings?
Yes: file meetings in folders and share a folder with your group.