Changelog
Systematic Transcribe
Systematic Transcribe 1.0
New
- Transcription that runs on your own machine — audio and video files, a folder at a time, a link, or a live recording. Nothing is uploaded: the model runs locally, online or offline
- Close to a hundred languages — pick one or let the model detect it, with the full range of Whisper models from tiny to large
- Speaker labels — the transcript says who spoke, and each speaker can be renamed once and renamed everywhere
- AI summaries — turn a finished transcript into a summary, meeting notes or a translation, through Claude or a local model on Ollama
- Global dictation — hold a shortcut anywhere on the system and the text arrives at the cursor
- Record from your phone — pair a phone once and send a recording to the desktop to transcribe, end-to-end encrypted
- Export — PDF, Word, HTML, Markdown, plain text, SRT, WebVTT, JSON, and a rule that exports every transcript the moment it finishes
- Six languages, both directions — English, Arabic, Spanish, French, Italian and Turkish, with Arabic reading right to left throughout
- An API and an agent skill — drive transcription from a script, the command line, or an AI agent