fun-voice-type for Openclaw

A macOS voice input plugin that uses Alibaba FunASR for real-time speech-to-text and multi-language translation directly into any input field.

yzwu2017
v2.0.0
Mar 21, 2026
0
727
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install fun-voice-type

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install fun-voice-type using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is fun-voice-type?

fun-voice-type is a powerful productivity enhancement tool designed for the macOS environment. By leveraging Alibaba's Cloud FunASR real-time speech recognition technology, it allows users to convert spoken words into text instantly. Unlike standard dictation tools, this skill within the Openclaw Skills framework integrates directly with your system, enabling you to "type" with your voice into any active cursor location, whether it's a browser, a code editor, or an IM client.

Beyond simple transcription, the skill supports robust multi-language translation capabilities. Users can switch target languages (such as English, Chinese, Japanese, and Korean) through a convenient menu bar icon, making it an essential utility for multilingual communication and fast-paced documentation.

fun-voice-type Use Cases

  • Rapidly inputting text when typing is inconvenient or slower than speaking.
  • Real-time transcription of notes during meetings or brainstorming sessions.
  • Instant translation of spoken phrases into multiple languages directly into chat or document windows.
  • Enhancing accessibility for users who prefer voice-over-keyboard interactions.

How fun-voice-type Works

  1. The user positions the mouse cursor in any standard macOS input field.
  2. By holding down the Right Option key, the skill activates the microphone and begins capturing audio.
  3. Real-time audio data is processed via the Alibaba DashScope FunASR API.
  4. Upon releasing the Right Option key, the recognized (and optionally translated) text is automatically injected into the current cursor position.
  5. Users can toggle translation settings and target languages via the system tray icon.

fun-voice-type Setup

1. Install System Dependencies

Install the necessary audio libraries and Python packages via Homebrew and pip:

brew install portaudio
pip install dashscope pynput pyaudio pystray

2. Configure API Access

Set up your Alibaba DashScope API Key as an environment variable to use Openclaw Skills securely:

export DASHSCOPE_API_KEY='your_api_key_here'

3. Execution

Run the script in the background to initialize the menu bar icon:

nohup python fun-voice-type.py > /dev/null 2>&1

fun-voice-type Data Schema & Taxonomy

The skill operates with a focused metadata structure and external API integration:

Component Description
Input Capture PCM audio stream captured via pyaudio.
API Integration Uses Alibaba DashScope (FunASR) for real-time WebSocket-based transcription.
Configuration Settings for target languages are managed through the pystray system tray menu.
Permissions Requires macOS Accessibility, Microphone, and Input Monitoring rights.

fun-voice-type Advanced Features

  • Global hotkey support using the Right Option key for non-intrusive activation.
  • Real-time multi-language translation (Chinese, English, Japanese, Korean).
  • System-wide compatibility with any macOS standard input control.
  • Low-latency processing powered by Alibaba's industrial-grade ASR models.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*