webchat-voice-proxy
Voice input and microphone button for OpenClaw WebChat Control UI. Adds a mic button to chat, records audio via browser MediaRecorder, transcribes locally via faster-whisper, and injects text into the conversation. Includes HTTPS/WSS reverse proxy, TLS cert management, and gateway hook for update safety. Fully local speech-to-text, no API costs. Real-time VU meter shows voice activity. Push-to-Talk (hold to speak) and Toggle mode (click start/stop), switchable via double-click. Keyboard shortcuts: Ctrl+Space PTT, Ctrl+Shift+M continuous recording. Localized UI (English, German, Chinese built-in, extensible). Keywords: voice input, microphone, WebChat, Control UI, speech to text, STT, local transcription, MediaRecorder, HTTPS proxy, voice button, mic button, push-to-talk, PTT, keyboard shortcut, i18n, localization.
Why use this skill?
Enable local voice input in OpenClaw WebChat with this proxy skill. Features real-time VU meters, push-to-talk, keyboard shortcuts, and full support for faster-whisper.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/neldar/webchat-voice-proxyWhat This Skill Does
The webchat-voice-proxy skill transforms your OpenClaw WebChat interface into a fully interactive, voice-enabled experience. It acts as a comprehensive local bridge between your browser’s microphone and a backend faster-whisper transcription engine. By injecting a responsive microphone button directly into your chat UI, the skill provides a seamless 'push-to-talk' or 'toggle' interface that captures audio via the browser's MediaRecorder API and transcribes it locally without relying on any external cloud APIs. This ensures complete data privacy and zero cost for transcription.
The skill features a sophisticated real-time VU meter that provides visual feedback by scaling and shadowing the microphone icon based on your voice activity. It includes robust local support for multiple languages (English, German, Chinese) and offers keyboard shortcuts like Ctrl+Space for instant push-to-talk or Ctrl+Shift+M for continuous recording sessions. Beyond the UI, it manages the underlying infrastructure by setting up an HTTPS/WSS reverse proxy, handling self-signed TLS certificate generation, and maintaining a systemd service to ensure the voice stack is reboot-safe and persistent.
Installation
Installation requires a running faster-whisper backend on port 18790. Once verified, use the command: clawhub install openclaw/skills/skills/neldar/webchat-voice-proxy. For standard local setups, execute bash scripts/deploy.sh. If you intend to access your OpenClaw instance from other devices on your network, use the environment variables: VOICE_HOST=<YOUR_LAN_IP> VOICE_HTTPS_PORT=8443 bash scripts/deploy.sh. This process automatically configures the systemd service and injects the necessary script hooks into your WebChat index file.
Use Cases
This skill is ideal for power users who prefer hands-free interaction with their AI agent. It is perfect for developers debugging voice-to-code workflows, accessibility-focused setups requiring voice input, or power-users who want to dictate long responses without using a keyboard. It turns a standard chat interface into a workstation-grade voice command center.
Example Prompts
- "OpenClaw, enable push-to-talk mode for my voice input so I can record short bursts of instruction."
- "Switch the microphone to toggle mode and confirm that the transcription language is set to German."
- "Start continuous recording with Ctrl+Shift+M and transcribe this meeting notes session directly into the chat."
Tips & Limitations
Note that the auto-generated TLS certificate will trigger a browser security warning upon your first visit; you must accept this manually. Because this relies on local hardware, ensure your browser has explicit microphone permissions for the specific domain used. If the VU meter remains static, verify that the openclaw-transcribe.service is active and that your default system audio input is correctly assigned.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-neldar-webchat-voice-proxy": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, code-execution
Related Skills
Faster Whisper Local Service
Skill by neldar
webchat-voice-full-stack
One-step full-stack installer for OpenClaw WebChat voice input with local speech-to-text. Deploys faster-whisper STT backend plus HTTPS/WSS WebChat proxy with mic button in one command. Push-to-Talk (hold to speak) and Toggle mode with keyboard shortcuts (Ctrl+Space PTT, Ctrl+Shift+M continuous recording). Real-time VU meter, localized UI (English, German, Chinese), interactive language selection during install. No recurring API costs, runs fully local after initial model download (~1.5 GB). Combines faster-whisper-local-service and webchat-voice-proxy. Keywords: voice input, microphone, WebChat, speech to text, STT, local transcription, whisper, full stack, one-click, voice button, push-to-talk, PTT, keyboard shortcut, i18n.