microjournal 0.1.3

A distraction-free terminal writing environment for Raspberry Pi

MicroJournal

MicroJournal turns a Raspberry Pi into a distraction-free terminal writing appliance.

It is built for headless use over SSH, so you do not need a GUI, a custom OS image, or any cloud account to get started.

If you want larger text, raise the font size in your terminal emulator and enable MicroJournal's large UI mode.

Install

uv tool install microjournal

If you want spaCy-powered NLP, install the optional NLP extra:

uv tool install "microjournal[nlp]"

If you want local voice dictation, install the optional voice extra:

uv tool install "microjournal[voice]"

The voice extra skips faster-whisper on Linux armv6l, because the upstream wheel stack is not available there.

Local development

uv run microjournal

Commands

• microjournal launches the editor
• microjournal setup runs first-time configuration
• microjournal config edits settings
• microjournal sync syncs notes
• microjournal export exports notes
• microjournal pack export writes a shareable visual pack JSON file
• microjournal pack load applies a visual pack from a JSON file
• microjournal doctor checks hardware and dependencies
• microjournal dictate runs push-to-talk local speech-to-text

Storage

• Notes are stored as Markdown in ~/MicroJournal/notes
• Config is stored in ~/.config/microjournal/config.toml
• Autosave is enabled by default and runs every 5 seconds

First run

1. Run microjournal setup
2. Choose your notes folder
3. Pick the display mode, boot behavior, sync placeholder, theme, and large UI preference
4. Launch microjournal

Editor shortcuts

• Ctrl+S save
• Ctrl+N new note
• Ctrl+F search notes
• Ctrl+P command palette
• Ctrl+T theme maker
• Ctrl+R view note stats
• Ctrl+E export note stats
• F8 toggle voice recording
• Ctrl+Q quit

NLP stats

• The footer shows live writing stats plus spaCy-powered NLP stats when the optional nlp extra is installed
• Note stats export creates a companion *.ner.md file with entities and summary stats
• If you install the optional nlp extra and the en_core_web_sm model is available, named-entity results are richer

Voice dictation

• The prototype uses the optional voice extra with faster-whisper plus microphone capture, all locally
• Dictated text is inserted into the editor at the cursor
• Voice mode is push-to-talk: press Voice or F8 to start, then press it again to stop

Accessibility

• Large UI mode increases spacing, sidebar width, and footer breathing room
• The app does not force terminal font size; that is still controlled by your terminal emulator
• Visual packs let you share a theme, palette, and display setup as a JSON file
• Use the command palette to open the in-app Theme Maker, tweak a pack, and export it for sharing

Roadmap

• Typing telemetry for sparklines and dashboards, focused on aggregated local writing data rather than raw keystrokes
• More visual pack presets for cyberdeck-style builds
• Better export/share flows for Reddit-ready screenshots and pack files