Skip to content
    ↑↓ select↵ openesc close
    English中文
    Ncastro878

    Say

    v1.0.0Notifications
    opencode-say

    Text-to-speech plugin for OpenCode - hear AI responses spoken aloud

    GitHub stars

    0

    Monthly installs

    7

    2 in 7 days

    Composite scoreSCORE

    16.7

    Multi-signal model

    Last commit

    5 months ago

    2026-03-08

    Install and configure

    opencode.json

    Writes to this project's opencode.json — applies to this repository only.

    opencode.json

    {
      "$schema": "https://opencode.ai/config.json",
      "plugin": ["opencode-say@1.0.0"]
    }

    opencode loads npm dependencies through its embedded runtime on startup and caches them locally — no manual global install needed.

    npm

    Text-to-speech plugin for OpenCode - hear AI responses spoken aloud.

    Uses macOS say command for text-to-speech. Perfect for hands-free coding with dictation software.

    Installation

    Add to your opencode.json:

    {
      "plugin": ["opencode-say"]
    }
    

    Or install globally in ~/.config/opencode/opencode.json.

    Usage

    Toggle Speech Mode

    Create a /say command by adding .opencode/command/say.md:

    ---
    description: Toggle speech mode on/off
    ---
    
    Use the `say_toggle` tool to toggle speech mode.
    

    Then type /say in OpenCode to toggle speech on/off.

    Available Tools

    The plugin exposes three tools that the AI can use:

    Tool Description
    say_toggle Toggle speech mode on/off
    say_config Configure voice, rate, and max length
    say_status Check current settings

    Configuration

    Settings are stored in .opencode/say-settings.json:

    {
      "enabled": true,
      "voice": "Samantha",
      "rate": 180,
      "maxLength": 500
    }
    
    Setting Type Default Description
    enabled boolean false Whether speech is enabled
    voice string system default macOS voice name
    rate number system default Words per minute
    maxLength number 500 Max characters before truncating

    Available Voices

    List available voices:

    say -v '?'
    

    Common high-quality voices:

    • Samantha - American female
    • Daniel - British male
    • Karen - Australian female
    • Moira - Irish female

    Requirements

    • macOS (uses the say command)
    • OpenCode v1.2.0+

    License

    MIT