Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
SAM (Software Automatic Mouth) is a text-to-speech engine that generates retro robotic audio in the style of the Commodore 64 era. It supports a persistent toggle mode where all agent responses are spoken aloud, plus one-off voice generation via the /sam command. Voice characteristics like pitch, speed, mouth, and throat are adjustable per-session.
Unlike cloud TTS services, SAM runs entirely offline with no API keys or network calls, producing its distinctive lo-fi robotic voice deterministically.
When to use it
- Adding robotic narration to automated notifications
- Generating retro-style voice replies in a chat session
- Testing phonetic pronunciation for C64-style audio projects
- Creating WAV voice clips for Discord or Telegram bots
- Toggling all agent responses into spoken robotic audio
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: /sam Hello there, what time is it?
- 1Extract text after /sam command: "Hello there, what time is it?"
- 2Run: node scripts/sam-tts-wrapper.js "Hello there, what time is it?" --output=/tmp/sam-001.wav --quiet
- 3Parse JSON output to confirm success and get output path
- 4Return generated WAV file as audio output
WAV audio file at /tmp/sam-001.wav playing the robotic voice saying "Hello there, what time is it?"