Speak Selection
Celina7 downloadsRight-click to read selected text aloud via Web Speech API. Auto-detects Chinese, Japanese, German, and English.
- Overview
- Scorecard
- Updates1
Right-click selected text to read it aloud via the Web Speech API. Auto-detects language — no configuration needed.
Features
- Right-click → "朗读选中文字/Speak Selection" on any text selection
- Command palette: trigger via the "朗读选中文字/Speak Selection" command or assign a hotkey
- Auto language detection: Chinese, Japanese, Korean, German, Russian, Arabic, Hindi, Thai, and more — detected by character set. German/English disambiguation via scoring (function words, prefixes, suffixes, noun capitalization)
- 23 supported languages as default/fallback (see below)
- Settings tab: toggle auto-detect, pick a default language, adjust speech rate
- Offline: uses built-in system voices, no API key or internet required
- Configurable speech rate: slider from 0.5x (half speed) to 2.0x (double speed), default 0.9x
Installation
Community Plugin Store
Search "Speak Selection" in Obsidian → Settings → Community Plugins.
Manual
Copy main.js and manifest.json into .obsidian/plugins/speak-selection/.
Usage
- Select text in the editor
- Right-click → 朗读选中文字, or run "朗读选中文字" from the command palette
- The plugin detects the language and speaks
Supported Languages
| Language | Code | Detection |
|---|---|---|
| 中文（普通话） | zh-CN | CJK unified ideographs |
| 中文（台灣） | zh-TW | — (default only) |
| 日本語 | ja-JP | Kana (hiragana/katakana) |
| 한국어 | ko-KR | Hangul syllables |
| Deutsch | de-DE | Umlauts/ß, or scoring-based detection |
| English (US) | en-US | Default when no other language matches |
| English (UK) | en-GB | — (default only) |
| Français | fr-FR | — (default only) |
| Español | es-ES | — (default only) |
| Italiano | it-IT | — (default only) |
| Português (PT) | pt-PT | — (default only) |
| Português (BR) | pt-BR | — (default only) |
| Русский | ru-RU | Cyrillic characters |
| العربية | ar-SA | Arabic script |
| हिन्दी | hi-IN | Devanagari script |
| ไทย | th-TH | Thai script |
| Nederlands | nl-NL | — (default only) |
| Polski | pl-PL | — (default only) |
| Türkçe | tr-TR | — (default only) |
| Svenska | sv-SE | — (default only) |
| Norsk (Bokmål) | nb-NO | — (default only) |
| Dansk | da-DK | — (default only) |
| Suomi | fi-FI | — (default only) |
| Ελληνικά | el-GR | — (default only) |
| Čeština | cs-CZ | — (default only) |
| Tiếng Việt | vi-VN | — (default only) |
Languages marked "— (default only)" are not auto-detected but can be set as the fallback language in settings.
Settings
- Auto-detect language: when enabled, the plugin automatically determines the language from the selected text. When disabled, the default language is always used.
- Default language: the fallback language when auto-detection is off or uncertain.
- Speech rate: slider from 0.5x to 2.0x (default 0.9).
Requirements
- Obsidian ≥ 0.15.0
- Desktop only (uses Web Speech API)
License
MIT
HealthExcellent
ReviewSatisfactory
About
Read selected text aloud locally via the Web Speech API with a right-click command and no API keys or network calls. Auto-detect Chinese (zh-CN), Japanese (ja-JP), German (de-DE) and English (en-US), disambiguate DE/EN by scoring function words, prefixes and suffixes, and select the best system voice.Details
Current version
1.0.1
Last updated
2 months ago
Created
2 months ago
Updates
1 release
Downloads
7
Compatible with
Obsidian 0.15.0+
Platforms
Desktop only
License
MIT