Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Covers ElevenLabs API workflows via direct HTTPS calls: text-to-speech, speech-to-speech voice conversion, realtime speech-to-text over WebSocket, and multi-voice dialogue generation. Includes auth patterns, voice/model discovery, and operational safety rules.
Direct HTTPS calls avoid SDK version drift and give full control over payloads, retries, and output format selection.
When to use it
- Convert a script to narrated audio with a specific voice
- Clone or convert a speaker's voice in an uploaded audio file
- Stream live transcription from a microphone feed
- Generate a two-character dialogue with distinct voices
- Look up available voice IDs before building a TTS pipeline
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: convert this blog post text to MP3 audio using a specific voice
- 1Reads elevenlabs-authentication.md to confirm API key header format
- 2Reads elevenlabs-voices-models.md to retrieve the target voice ID and a compatible model ID
- 3Reads elevenlabs-text-to-speech.md to construct the POST /v1/text-to-speech/{voice_id} request with the chosen output format
- 4Applies retry-with-backoff logic and caches voice/model IDs for subsequent calls
- 5Returns the audio file and the endpoint checklist used
MP3 audio file of the blog post narrated in the selected voice
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.