Scribe

On-device transcription, with speakers labeled.

Scribe turns a recording — a meeting, an interview, a lecture, a voice memo — into accurate text that runs entirely on your device. Nothing is uploaded to a server. It also does speaker diarization: it works out who said what and labels each turn, so a two-person interview reads like a transcript instead of a wall of text.

What it does

• Private by design — speech recognition runs locally with an on-device model; audio never leaves the phone.
• Speaker labels — diarization separates voices and tags each segment.
• Import or record — transcribe an existing audio file or capture new audio in the app.
• Export the finished transcript as text or PDF and share it.