Inworld AI

Inworld AI MCP Connector for Claude

F

Power your AI agents with Inworld's lifelike voices, voice cloning, and advanced character orchestration routers.

19 tools Official Updated Jun 28, 2026 Official Vinkius Partner

Connect Inworld AI to your agent to unlock advanced multimodal capabilities. This server provides a comprehensive suite of tools for high-fidelity voice synthesis, custom voice design, and intelligent character routing.

What you can do

  • Voice Synthesis (TTS) — Generate high-quality speech synchronously or via streaming using advanced models like inworld-tts-2.
  • Voice Cloning & Design — Clone voices from audio samples using clone_voice or describe a voice with text prompts using design_voice to create unique personas.
  • Character Orchestration — Manage AI routers with create_router and chat_completions to build complex, context-aware conversational experiences.
  • Voice Library Management — List, update, and organize your workspace's voice assets and published models.
  • Speech-to-Text (STT) — Transcribe audio files directly into text for seamless agent processing.

How it works

  1. Subscribe to this server
  2. Enter your Inworld API Key and API Secret from the Inworld Studio
  3. Start generating speech and managing AI characters from Claude, Cursor, or any MCP client

Who is this for?

  • Game Developers — Create dynamic, voiced NPCs with unique personalities and real-time responses.
  • Content Creators — Automate voiceovers and character dialogue using custom-designed or cloned voices.
  • AI Engineers — Build sophisticated agentic workflows that require advanced routing and multimodal interaction.
text-to-speechvoice-cloningai-charactersconversational-aispeech-synthesis

19 tools expose this connector's capabilities to your AI agent.

chat_completions

Generate chat completions via LLM Router

clone_voice

Clone a voice from audio samples

create_realtime_call

Create a WebRTC realtime call

create_router

Create an LLM Router

delete_router

Delete an LLM Router

delete_voice

Delete a voice

design_voice

Design a voice from a text prompt

get_router

Get details for a specific LLM Router

get_voice

Get details for a specific voice

list_models

List available LLM models

list_routers

List all LLM Routers

list_tts_voices

Use list_voices instead for newer Voice API. List TTS voices (Deprecated)

list_voices

List voices in the workspace

publish_voice

Publish a draft/preview voice

synthesize_speech_stream

Synthesize speech via streaming

synthesize_speech_sync

Synthesize speech synchronously

transcribe_audio

Transcribe audio to text (Sync)

update_router

Update an LLM Router

update_voice

Update a voice

See how to talk to your AI agent using Inworld AI.

List all the voices available in my Inworld workspace.

I've retrieved your voice library. You have 12 voices available, including 'Narrator Alpha' (ID: v-123) and 'Heroic Guide' (ID: v-456). Would you like to see details for a specific one?

Synthesize the text 'Welcome to the future of AI' using voice ID 'v-123' and the expressive delivery mode.

Processing synthesis with `synthesize_speech_sync`... The audio has been generated successfully using the `EXPRESSIVE` delivery mode. You can now play or download the result.

Design a new voice that sounds like a mysterious, deep-voiced wizard for my game.

I'm using `design_voice` with your description. I've generated a preview for a 'Mysterious, deep-voiced wizard'. Would you like to hear the preview or publish this voice to your workspace?

You can use the `design_voice` tool. Simply provide a prompt like 'Warm, friendly male voice' and a preview text. The tool will generate voice options that you can later publish to your library.

Related Connectors