Eachlabs Voice Audio

This is a unified interface to EachLabs' voice and audio models, covering text-to-speech (ElevenLabs, Kokoro, Stable Audio), speech-to-text (Whisper variants, ElevenLabs Scribe), and voice conversion (RVC v2, XTTS). You get 30+ models through one API with a consistent prediction flow: check the model schema, POST your request, poll for results. The main draw is breadth. Need speaker diarization? It's there. Want to merge dubbed audio back into video? FFmpeg endpoint included. The docs are clear about checking each model's input schema before use, which matters because parameters vary wildly between models. Good for building voice workflows without managing multiple vendor integrations, though you're still dealing with async prediction polling for everything.

Install to Claude Code

npx -y skills add eachlabs/skills --skill eachlabs-voice-audio --agent claude-code