Chat or go hands-free with voice mode — put your phone in your pocket and keep talking via Bluetooth headphones. Your server, your providers, your rules.
Have OpenClaw? Just tell your agent:
Install the Clack voice relay skill from https://github.com/fbn3799/clack-skill and set it up
Choose speech-to-text and text-to-speech providers separately for voice mode. Mix on-device with cloud. Save money without sacrificing quality.
Example: Free on-device transcription + premium ElevenLabs voices = great quality, zero STT costs.
Message your OpenClaw AI by text or switch to hands-free voice mode whenever you need it.
Speak naturally and get spoken responses. Put your phone in your pocket and keep talking via Bluetooth headphones. Automatic silence detection, interrupt anytime.
Full text chat with streaming responses. Create, rename, and switch between multiple conversations. Voice and text in the same thread.
Built-in voice support for ElevenLabs, OpenAI, and Deepgram. Or use system voices for free on-device speech.
Create, rename, and switch between multiple conversations. Each thread keeps its own history, synced from your server.
Mute and unmute with play/pause. Full remote command center integration for hands-free conversations.
Use free on-device STT and only pay for premium TTS — or go fully local for zero API costs. You decide.
Clack is private by design. No data leaves your infrastructure.
Voice audio goes to your server — and only to the providers you choose. The developer never sees or touches your data.
No analytics, no tracking, no usage data. The app stores settings locally and that's it.
Constant-time HMAC token verification. One-time pairing codes with 5-minute expiry and rate limiting.
Connect via Domain with SSL or Tailscale — the app enforces encrypted connections. No unencrypted public access.
Server, protocol, and app — fully auditable. Verify the security claims yourself.
No sign-up, no email, no personal information. Pair with a code and start talking.
Self-hosted means you own the infrastructure. Here's how to set it up.
Already have OpenClaw? Just tell your agent:
Install the Clack voice relay skill from https://github.com/fbn3799/clack-skill and set it up
Your agent handles cloning, dependencies, and service setup automatically.
Or set it up manually:
Clone the repo and run the interactive setup script. It installs dependencies, prompts for API keys, and configures the service.
git clone https://github.com/fbn3799/clack-skill.git ~/.openclaw/skills/clack
sudo bash ~/.openclaw/skills/clack/scripts/setup.sh
No API keys? No problem — on-device speech works without any.
The setup script asks you to choose:
Domain (SSL): Point a DNS A record to your server — SSL is configured automatically via Caddy. Free domains via DuckDNS work too.
Tailscale: Install on server + phone, use the Tailscale IP. No domain or SSL setup needed.
Get Clack on Google Play or the App Store. Or build your own client using the open-source server.
Enter your domain or Tailscale IP in Settings → Server.
Domain mode: Tap "Pair with Server" and enter the 6-character code from setup.
Tailscale: No pairing needed — just connect directly.
After setup, manage your server with the clack command: clack status · clack pair · clack logs
Server source: github.com/fbn3799/clack-skill
Text or voice — your AI, your rules. No subscription, no middleman.