Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Security scan fixes
What This Skill Does
Generates expressive text-to-speech audio using the ElevenLabs v3 API with inline emotional audio tags. Supports 70+ languages and produces WhatsApp-compatible voice messages via ffmpeg conversion. Emotional tone, pacing, and delivery are controlled directly in the text using bracket tags like [excited], [whispers], or [pause].
ElevenLabs v3's audio tag system controls emotion and pacing inline, removing the need for separate audio editing after generation.
When to use it
- Sending emotional voice messages on WhatsApp
- Narrating horror or suspense stories with audio tags
- Generating multilingual voice content in Hebrew, Spanish, or French
- Building multi-speaker podcast segments from a script
- Converting celebration announcements into expressive audio clips
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Send a WhatsApp voice message saying I got into college with excitement
- 1Compose text with emotional tags: '[excited] I got in! [laughs] [happy] This changes everything! [pause]'
- 2Call tts tool with the tagged text and channel=whatsapp
- 3Run ffmpeg to convert the returned MP3 to Opus format
- 4Call message tool with the Opus file path, asVoice=true, and target phone number
WhatsApp voice message delivered with excited, laughing tone
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.