Speaker names, added by you
Edit a transcript segment to add or change its speaker name. It works the same on Mac and Windows.
JesRecap no longer detects speakers automatically, on either platform. The app does not download or run a speaker model. You name speakers by hand in the transcript editor. Names you typed, and names saved by earlier builds, are kept and stay editable.
Why this page says "diarization"
Speaker diarization is the technical name for working out who was talking, and when. Earlier preview builds did this automatically on Mac, with a local model. That feature has been removed. We kept this address so old links still land somewhere useful.
What JesRecap offers now is simpler. The transcript is written after you press Stop, in time-stamped segments. You decide who said what, and type the name.
How it works in JesRecap
Manual, local, and the same on Mac and Windows.
Edit a segment, type a name
Press Edit on any transcript segment. Alongside the text there is a speaker field. Type a name, save, and it appears on that segment.
No automatic detection
JesRecap does not guess who is speaking. No speaker model is downloaded or run, and nothing extra happens after Stop beyond transcription.
Your names are kept
Names you typed stay with the meeting. Names saved by earlier builds, including automatic ones, are kept too, and you can still edit them.
Nothing leaves the device
Names are stored with the transcript on your disk. They go somewhere else only if you export, share or paste them.
Carried into exports
Speaker names appear in the transcript view and in what you export: Markdown notes, plain text, the JSON archive, and SRT/VTT captions.
Your side on its own track
Your microphone is still recorded to its own file. Export the microphone-only WAV when you need to hear your side alone.
Separate tracks still help
JesRecap records your microphone and the meeting audio to separate files, then mixes a playback track after the meeting. The transcript is made from that mix, so it does not mark your lines for you. The separate files are still useful when you are naming speakers by hand.
- You can check the source. Export microphone-only or system-audio-only WAV to hear a single side of the conversation on its own.
- Interruptions survive in the audio. When you and someone else talk over each other, the two voices sit on two different files.
- Retranscribe later. Both tracks stay on disk, so an old meeting can be transcribed again with a newer model.
Honest limits
- Naming takes your time. Long meetings with many voices need more editing. There is no shortcut that names every segment at once.
- Segments are time-based. A segment can contain more than one voice when people speak quickly back and forth. Its name is whatever you type.
- Not evidence. A name in a transcript is an annotation, not proof of who said what in a dispute. Keep the audio, which is on your disk, and listen.
Speaker names, answered
Does JesRecap still detect speakers automatically?
No. Automatic speaker detection has been removed on Mac and Windows. The current app does not download or run a speaker model. You add speaker names by hand in the transcript editor.
What happens to speaker names in meetings I already labeled?
They stay. Names saved by an earlier build, and names you typed yourself, are kept with the meeting and remain editable.
How do I name a speaker?
Open the meeting, press Edit on a transcript segment, and type a name in the speaker field. Save, and the name appears on that segment.
Is my own voice labeled for me?
Not in the transcript. Your microphone is still recorded to its own file, so you can export microphone-only audio and hear your side on its own, but the transcript is made from the mixed track and you name speakers yourself.
Do speaker names appear in exports?
Yes. Names you add are included in Markdown notes, plain text, the JSON archive, and SRT/VTT caption files.
Name who said what, privately
Manual speaker names, a separate track for your own voice, and no audio leaving your machine. From €34.95 once.