Voice Audio Mcp

Wraps Kokoro TTS and Whisper STT behind three clean tools: text to speech conversion, voice listing, and audio transcription. Built by MEOK AI Labs with EU AI Act compliance hooks baked in. Useful when you need Claude to generate spoken audio from text, switch between different voice profiles, or convert audio files back into text transcripts. Runs over stdio, installs via pip or uvx. The compliance layer adds watermarking and risk management scaffolding if you're shipping customer-facing voice features in regulated contexts. Straightforward Python implementation without external service dependencies for the core transcription and synthesis paths.