Elevenlabs

This converts documents and text to audio using ElevenLabs TTS, with two modes: single-voice narration and two-host podcast generation. The podcast workflow is the standout feature. It extracts text from PDFs or docs, lets you generate a conversational script as JSON (with speaker turns), then synthesizes it with two different voices. Handles chunking and concatenation automatically via ffmpeg. You can also just narrate a document straight with one voice. Requires an ElevenLabs API key and works best if you list available voices first so users can pick what they want. Good for turning written content into listenable audio without manual recording.

Install to Claude Code

npx -y skills add sanjay3290/ai-skills --skill elevenlabs --agent claude-code