FishSpeech TTS MCP

Example Use Cases

• 1

  Convert generated text into spoken audio using FishSpeech for voice interactions or audio responses.

• 2

  Create and store voice references (voice cloning) by uploading example audio and corresponding transcripts for later reuse.

• 3

  An MCP-compatible TTS backend to integrate with Dive or other MCP-enabled LLM frontends via a simple npx-installable server.

Description

This MCP server connects to an Optivus backend via Socket.IO to convert text to speech and to save voice references for voice cloning. It integrates with MCP-capable LLM frontends such as Dive and can be installed via npx for easy deployment. The project supports saving voice samples in a structured directory and handles reconnection and error scenarios automatically. It is an open-source, MIT-licensed tool intended to add TTS capabilities to agent workflows.