About
Plainsong is a local-first dictation and meeting capture app.
Voice, faithfully written.
Plainsong is being built for people who turn spoken context into written work. The release path focuses on fast desktop dictation, local Whisper transcription, meeting capture, speaker labeling, and exportable records.
The product direction is privacy-aware by default. Core capture and transcript review are designed around local processing, while cloud services stay optional for teams that need them.
The current public launch page is intentionally careful about claims. It shows the direction of the product, the open-source release track, and the local-first workflow without pretending that every build path is final. That boundary matters because dictation and meeting notes can contain sensitive work, customer context, or private commitments.
Who it is for
Plainsong is meant for people who need spoken work to become usable records: support notes, planning calls, interviews, customer follow-up, personal drafting, and review notes that need to survive beyond the moment they were said. The app keeps capture, transcript review, labels, and export paths close together so the record can be checked later.
The constraint that shapes it
The project fits the same pattern as Jonathan's other public tools: keep private workflows clear, make evidence easier to inspect, and avoid vague AI claims. Local processing is not a slogan here; it is the product constraint that shapes how recording, transcription, review, and handoff are described.
That same constraint shapes the launch messaging. Plainsong should be easy to explain without overpromising: capture audio, turn it into text, let the user review the record, and make export decisions explicit. Anything involving sync, teams, accounts, or hosted transcript storage should be named plainly before it ships.
The app is maintained by Jonathan R. Reed. For product, privacy, or release questions, use the Plainsong contact page.
Last updated