Dictate lets you record your voice, transcribe it locally with whisper.cpp, and turn it straight into a Joplin note or to-do. Audio never leaves your machine.
Features
- One-click dictation panel with Pause / Resume and Cancel
- Fully local transcription via
whisper.cpp(whisper-cli) — private by design - Optional transcript polishing (grammar, punctuation, filler cleanup) via any OpenAI-compatible LLM endpoint
- Transcribe an existing WAV file instead of recording
Requirements
- Linux with PipeWire (
pw-record) for audio capture - whisper.cpp built locally with a ggml model (e.g.
ggml-small.en.bin) - Joplin 3.6+
Install
- Plugin manager: Tools → Options → Plugins → search Dictate (once the bot has indexed it)
- Manual: download the
.jplfrom the v1.0.0 release - npm:
joplin-plugin-dictate
Source & issues
github.com/NormG/joplin-dictate-plugin · MIT licensed
This is my first Joplin plugin 
— feedback and bug reports very welcome! It's currently Linux-only because audio capture uses PipeWire's pw-record; if there's interest I'm open to adding other capture backends.
(Screenshots below)