voice-ui

What This Skill Does

The voice-ui skill is a revolutionary self-evolving interface for the OpenClaw ecosystem. It transforms your desktop interaction by providing a conversational, expressive robot UI that performs both text-to-speech and speech-to-text functions using Whisper and TTS technologies. Unlike static assistants, voice-ui leverages a feedback loop where the AI can modify its own codebase, style, and functionality in real-time based on your vocal instructions. By utilizing a Node.js backend to facilitate file system changes and automatic Git versioning, it ensures that every enhancement you request is safely tracked and immediately applied to the browser-based environment.

Installation

To integrate this skill, ensure you have the OpenClaw environment initialized. First, install the skill by running: clawhub install openclaw/skills/skills/yukihamada/voice-ui. Navigate to the skill directory, typically located at <workspace>/skills/voice-ui, and execute npm install to resolve dependencies. You must provide your OPENAI_API_KEY in your environment variables to enable the underlying speech processing engine. Finally, run the provided ./start.sh script and navigate to http://localhost:8765 to engage with your voice-activated assistant.

Use Cases

Voice-ui is designed for users who want a hands-free, dynamic interaction layer for their workflows. It is perfect for developers who want to prototype UI components by simply describing them; visual designers who need to tweak aesthetics without opening IDEs; and power users looking for a personalized AI companion that changes its look and feel to match their current mood or productivity needs. Because it records all changes to Git, it is also an excellent tool for experimenting with UI states without fear of losing your original configuration.

Example Prompts

1. "I want the assistant to look more modern. Can you change the background to a gradient purple and round out the button corners?"
2. "Add a small dashboard widget in the top right corner that displays my current system CPU usage."
3. "The robot interface feels a bit too large. Could you reduce the scale by 20% and make the response text easier to read?"

Tips & Limitations

To get the most out of voice-ui, be specific with your design requests—mention colors, sizes, or layout changes clearly. Keep in mind that since the AI modifies the code directly, very complex logic changes might occasionally require a manual review of the codebase. Always ensure your Git environment is properly configured so the auto-commit feature can track your modifications effectively. Note that this skill requires constant network access for the OpenAI services to function, and browser permissions must be granted to allow the application access to your microphone.