Aller au contenu
TakoAPI
Retour aux skills

webchat-voice-full-stack

par neldar

Web & Frontend Development
554Téléchargements rapportés par clawskills.sh
4Étoiles
0Vues
0 mention J'aimeConnecte-toi pour aimer ce skill
—Pas encore noté
Ta note

Résumé

One-step full-stack installer for OpenClaw WebChat voice input with local speech-to-text.

Description

Ce que fait ce skill

Meta-installer that sets up voice input for OpenClaw WebChat by orchestrating three sub-skills in sequence: a local speech-to-text backend (faster-whisper), an HTTPS/WSS reverse proxy, and a voice UI with mic controls. Verifies SHA256 checksums of all sub-skill scripts before execution. No recurring API costs after the initial model download.

Handles ordering, checksum verification, and environment variable passthrough across three interdependent skills so users avoid manual coordination errors.

Quand l'utiliser

  • Adding voice input to a self-hosted OpenClaw WebChat instance
  • Setting up local transcription without sending audio to external APIs
  • Deploying push-to-talk shortcuts for hands-free chat input
  • Installing voice UI for multilingual teams (EN/DE/ZH)
  • Running a one-command setup instead of configuring three separate skills manually

Exemple de workflow

ENTRÉE

User runs: bash scripts/deploy.sh

AGENT
  1. 1

    Verifies SHA256 checksums of all sub-skill scripts against checksums.sha256, aborts if any mismatch

  2. 2

    Calls faster-whisper-local-service deploy: creates Python venv, installs faster-whisper, writes transcribe-server.py, registers systemd user service

  3. 3

    Calls webchat-https-proxy deploy: copies https-server.py, appends HTTPS origin to gateway allowedOrigins, generates self-signed TLS cert, registers systemd user service

  4. 4

    Calls webchat-voice-gui deploy: copies voice-input.js, injects script tag into Control UI, installs gateway startup hook, prompts for language selection

SORTIE

Three systemd user services running, voice mic button visible in WebChat UI with push-to-talk and VU meter active

Prérequis

  • Python 3.10+ installed on the system
  • GStreamer (gst-launch-1.0) installed via OS packages
  • Internet access on first run for Whisper model download (~1.5 GB)

Installation

clawhub install neldar/webchat-voice-full-stack

Exécute la commande CLI, ou colle l'instruction dans Claude Code, Cursor, Codex ou l'agent de code que tu utilises.

Liens