TTS Highlight

Read your Obsidian notes aloud with real-time word-by-word highlighting. Uses your system's built-in text-to-speech voices via the browser SpeechSynthesis API — no cloud services, no API keys, completely offline.

Features

• Word-by-word highlighting as text is spoken, in all editor modes:
  • Source mode (CM6 decorations)
  • Live Preview (including inside rendered tables)
  • Reading mode (CSS Custom Highlight API with <mark> fallback)
• Read full note, from cursor, or selection only
• Pause, resume, and stop via commands, ribbon icon, or status bar controls
• Configurable voice, rate, pitch, and volume
• Custom highlight color or use your vault's accent color
• Auto-scroll to keep the current word visible
• Smart markdown stripping — headings, bold/italic markers, link syntax, code fences, frontmatter, and table formatting are stripped before speaking, with a character-level position map back to the editor
• Large document support — text is chunked at sentence boundaries to avoid browser speech cutoffs
• Edge case handling — stops on document edit, stops on note switch, handles Safari quirks (missing charLength, end event not firing after cancel)

Commands

Command	Description
Read aloud	Read the entire note from the beginning
Read from cursor	Read from the current cursor position to the end
Read selection aloud	Read only the selected text
Pause / Resume	Toggle pause (starts reading if idle)
Stop	Stop playback and clear highlighting

No default hotkeys are assigned to avoid conflicts. Bind them in Settings > Hotkeys by searching for "TTS Highlight".

Settings

Setting	Description	Default
Voice	Select from available system voices	System default
Rate	Speech speed (0.5x - 2.0x)	1.0
Pitch	Voice pitch (0.5 - 2.0)	1.0
Volume	Playback volume (0.0 - 1.0)	1.0
Highlight color	CSS color for the highlighted word	Accent color at 35% opacity
Auto-scroll	Scroll to keep the highlighted word visible	On
Chunk size	Max characters per speech chunk	5000

The settings tab also includes a Hotkeys section with quick access to configure keyboard shortcuts for each command.

Installation

From Community Plugins

1. Open Settings > Community plugins
2. Search for "TTS Highlight"
3. Click Install, then Enable

Manual Installation

1. Download main.js, styles.css, and manifest.json from the latest release
2. Create a folder tts-highlight in your vault's .obsidian/plugins/ directory
3. Copy the three files into that folder
4. Enable the plugin in Settings > Community plugins

Known Limitations

• Desktop only. The SpeechSynthesisUtterance boundary event (which provides word positions) does not fire on Android. The plugin is marked isDesktopOnly in the manifest.
• Voice availability varies by OS. macOS includes high-quality voices out of the box. Windows and Linux may have fewer options — install additional voices through your OS settings.
• Chrome long-utterance cutoff. Chrome silently stops speaking after ~15 seconds of continuous speech. The plugin mitigates this by splitting text into chunks at sentence boundaries. If you experience cutoffs, try reducing the chunk size in settings.
• Safari quirks. Safari does not report charLength in boundary events and does not fire the end event after cancel(). The plugin includes workarounds for both.

Development

# Install dependencies
npm install

# Build (watch mode)
npm run dev

# Production build
npm run build

License

MIT