Pronounce Selected Text (Alt+C)

Update (v1.2.2): Fixes the October 2026 Microsoft Edge TTS outage (HTTP 403 with older edge-tts 7.2.8). Neural voices are restored via bundled edge-tts 7.3.1. Cache keys now include the synthesis engine, so a temporary Google TTS fallback cannot overwrite a neural voice cache entry. After updating: restart Anki. If some phrases still sound lower quality, delete files in the add-on user_files/cache/ folder and pronounce them again.

Instant, ultra-realistic text-to-speech pronunciation of highlighted text or card example sentences in Anki upon pressing Alt+C (configurable). Powered by Microsoft Azure / Edge Neural voices with automatic on-the-fly language detection.

Key Features

• Dual-Mode Pronunciation:
  • Selection Mode: Highlight any word, clause, idiom, or full sentence across the Card Reviewer, Note Editor, or Card Browser and press Alt+C to hear it spoken instantly.
  • Fallback Field Mode: Press Alt+C without selecting any text, and the add-on automatically pronounces a configured note field (default: ItalianExample; set any field name in Config).
  • Smart Conjugation Matching: On verb conjugation cards where examples are organized per grammatical person (Io, Tu, Loro), it dynamically detects the active card template and speaks the matching person's example (e.g. Loro example).
• Automatic Language Detection: Pure-Python, zero-dependency detector identifies the language of the target text (Italian, French, Spanish, German, Russian, English, Portuguese, Japanese, Chinese, etc.) and routes to the appropriate native voice.
• Dedicated Native Monolingual Voices: Defaults to dedicated monolingual neural voices (it-IT-ElsaNeural, fr-FR-DeniseNeural, es-ES-ElviraNeural, de-DE-KatjaNeural, ru-RU-SvetlanaNeural, en-US-JennyNeural) to reduce English phonetic leakage on shared vocabulary.
• Dynamic SSML Locale Integrity: Root <speak xml:lang='...'> dynamically matches the detected voice locale for authentic native intonation.
• Multi-Layout Keyboard Support (English & Russian): Built-in physical key code and Cyrillic mapping ensures Alt+C triggers smoothly whether your active system keyboard layout is English or Russian.
• Zero-Restart Live Configuration: Adjust shortcuts, volume, or fallback fields in Tools → Add-ons → Config and changes apply immediately without restarting Anki.
• Instant Local Caching (<1 ms): Synthesizes new phrases in ~300 ms and caches them locally. Repeat plays are instant. Cache slots are separated by engine (edge vs fallback) so degraded fallback audio cannot replace neural audio.
• Audio Preemption: Rapid keypresses cleanly terminate previous playback so new audio begins immediately.
• 100% Standalone: Fully self-contained with bundled dependencies. No external add-ons, accounts, or API keys required.

Supported Languages & Dedicated Voices

• Italian: it-IT-ElsaNeural
• French: fr-FR-DeniseNeural
• Spanish: es-ES-ElviraNeural
• German: de-DE-KatjaNeural
• Russian: ru-RU-SvetlanaNeural
• English: en-US-JennyNeural
• Portuguese: pt-BR-FranciscaNeural
• Japanese: ja-JP-NanamiNeural
• Chinese: zh-CN-XiaoxiaoNeural
• (Plus Greek, Arabic, Hebrew, and other standard languages)

Prerequisites (Linux)

Ensure mpv is installed on your Linux system:

• Arch Linux / Omarchy: sudo pacman -S mpv
• Ubuntu / Debian: sudo apt install mpv
• Fedora: sudo dnf install mpv

Configuration

In Anki, navigate to Tools → Add-ons → select Pronounce Selected Text → click Config:

{
  "shortcut": "Alt+C",
  "volume": 140,
  "audio_output": "pipewire,pulse",
  "speed": 1.0,
  "show_tooltip": false,
  "fallback_field": "ItalianExample",
  "debug_log": true
}

Support / issues: GitHub Issues
Source: https://github.com/argrig666/anki-pronounce-selected