How to Use the Inworld AI MCP in LangChain

Q: How do I handle Inworld AI audio streaming inside a LangChain runnable?

Use synthesizespeechstream inside your custom runnable block to yield chunks. LangChain handles generators naturally, so you can stream the raw audio bytes directly to your frontend.

Q: Can a LangChain agent dynamically clone voices based on user inputs?

Yes, the agent can call clonevoice using audio sample URLs provided in the input. Once the voice is ready, the agent passes the new voice ID to synthesizespeechsync in the next step.

Q: How does this MCP Server handle router updates during a LangChain run?

Your agent can call updaterouter mid-conversation to adjust character prompts or models dynamically. This allows the agent to change its behavior based on previous chain outputs.

Q: What is the best way to trace Inworld AI latency in LangChain?

Turn on LangSmith tracing to monitor every call to createrealtimecall or chatcompletions. You will see the exact millisecond breakdown of the audio synthesis and WebRTC connection times.

Build multi-step voice agent pipelines in LangChain using lifelike vocal cloning and dynamic router tools.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Inworld AI MCP to LangChain

Create your Vinkius account to connect Inworld AI to LangChain and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Inworld AI with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain voice cloning with LangChain agent workflows

Running `clone_voice` or `design_voice` dynamically within your conversational chains replaces static audio files with real-time vocal profiles. This MCP Server lets your LangChain agents execute these tools on the fly, passing the resulting voice ID straight into the next step of your chain. Because every step is a clean link in your run, you can immediately feed that voice ID into `synthesize_speech_stream` to generate audio. LangSmith traces the entire latency profile, showing you exactly how fast your audio packets stream back.

Route conversations dynamically using LLM routers

Calling `create_router` directly from your agentic loop offloads heavy character context routing to Inworld. LangChain agents excel at choosing the right path, but managing complex character states is tough without this dedicated MCP Server routing layer. Your agent can inspect the active paths with `list_routers` and quickly switch contexts using `update_router`. By offloading the character logic to the router, you cut down on prompt tokens and avoid bloating system messages.

Real-time WebRTC audio loops in your chains

Triggering `create_realtime_call` from your LangChain pipeline spins up a WebRTC session so users can talk directly to your agent. Static text-to-speech doesn't cut it when you need live voice interactions, making an MCP connection essential. While the call is active, the agent uses `transcribe_audio` to convert incoming voice data on the fly. You keep the conversational loop tight without breaking your chain's stateful memory.

Setup guide

Set up Inworld AI MCP in LangChain

Prerequisites

Python 3.10+ installed
langchain-mcp-adapters + langgraph packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install langchain-mcp-adapters langgraph langchain-openai. The MCP adapters package converts MCP tools into native LangChain BaseTool objects.
2

Connect via HTTP transport
Use MultiServerMCPClient with "transport": "http" pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create a ReAct agent
Pass the discovered tools to create_react_agent() from LangGraph. The agent automatically routes Inworld AI tool calls through the MCP protocol.
4

Run with any LLM
Swap ChatOpenAI for ChatAnthropic, ChatGoogleGenerativeAI, or any LangChain-compatible model. The MCP tools work identically across all providers.

agent.py

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_openai import ChatOpenAI

async with MultiServerMCPClient({
    "inworld-ai-mcp": {
        "transport": "http",
        "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
    }
}) as client:
    tools = client.get_tools()

    agent = create_react_agent(
        ChatOpenAI(model="gpt-4o"),
        tools,
    )
    result = await agent.ainvoke({
        "messages": "List recent Inworld AI transactions"
    })
    print(result["messages"][-1].content)

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Inworld AI. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Inworld AI now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Inworld AI MCP in LangChain

Use `synthesize_speech_stream` inside your custom runnable block to yield chunks. LangChain handles generators naturally, so you can stream the raw audio bytes directly to your frontend.

Yes, the agent can call `clone_voice` using audio sample URLs provided in the input. Once the voice is ready, the agent passes the new voice ID to `synthesize_speech_sync` in the next step.

Your agent can call `update_router` mid-conversation to adjust character prompts or models dynamically. This allows the agent to change its behavior based on previous chain outputs.

Turn on LangSmith tracing to monitor every call to `create_realtime_call` or `chat_completions`. You will see the exact millisecond breakdown of the audio synthesis and WebRTC connection times.

Your cloned audio samples and voice profiles are stored securely in your private workspace on the Vinkius sandbox. All transmissions use encrypted HTTPS endpoints, ensuring your raw audio files never leak during execution.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript