How to Use the Cartesia (Voice AI) MCP in OpenAI Agents SDK

Q: Can the OpenAI Agents SDK stream audio from Cartesia (Voice AI)?

Yes. You use the ttssse tool for Server-Sent Events. This streams the generated speech chunks back to your application before the full sentence finishes processing.

Q: Does this integration support batch transcription?

It does. The sttbatch tool accepts audio files and returns text transcripts. Your OpenAI agent handles the resulting text for downstream analysis or summarization.

Q: How do I check credit usage for Cartesia (Voice AI) via OpenAI Agents SDK?

Call the getusagecredits tool. Your agent pulls the exact billing stats. Set cacheToolsList=True in your setup so the agent knows this tool exists without pinging the server first.

Connect Cartesia's sonic models to your OpenAI Agents SDK workflow. Generate speech, clone voices, and transcribe audio in production.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Cartesia (Voice AI) MCP to OpenAI Agents SDK

Create your Vinkius account to connect Cartesia (Voice AI) to OpenAI Agents SDK 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 Cartesia (Voice AI) with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Cartesia MCP Server for OpenAI Agents SDK

Your multi-agent system needs a voice. This MCP server exposes the `tts_bytes` and `tts_sse` tools straight to your OpenAI models. You pass text strings, and the agent returns raw audio bytes ready for playback in your frontend application. Need a specific sound? The `clone_voice` tool takes a five-second audio clip and builds a custom profile. You pair that with `localize_voice` to make your new voice speak different languages, all while the OpenAI guardrails validate every API call before it executes.

Fix audio gaps programmatically

Splicing audio usually takes a human editor. We give your agent the `infill_bytes` tool to connect two disjointed audio segments smoothly. It generates the missing speech right in the middle, matching the surrounding acoustic environment. You also get `voice_changer_bytes`. Pass an existing audio file through this tool, and the agent swaps the speaker's identity while keeping the original intonation intact. Your tracing dashboard logs exactly how many credits you burn via `get_usage_credits`.

Control pronunciation at scale

Industry terms trip up standard text-to-speech engines. The `create_pronunciation_dict` tool fixes this by letting your agent define exact phonetic spellings for tricky words. You update these rules on the fly with `update_pronunciation_dict`. When you deploy voice agents, you must know what they say. The `list_agent_calls` tool pulls full transcripts for any specific agent. You monitor the data, and your OpenAI setup handles the handoffs between your text and voice workers.

Setup guide

Set up Cartesia (Voice AI) MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Cartesia (Voice AI) tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Cartesia (Voice AI) tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Cartesia (Voice AI) tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Cartesia (Voice AI) Agent",
            instructions="You have access to Cartesia (Voice AI) tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

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 Cartesia. 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 Cartesia (Voice 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 Cartesia (Voice AI) MCP in OpenAI Agents SDK

Run pip install openai-agents first. Then initialize MCPServerStreamableHttp with your Vinkius endpoint URL. Pass that directly to your Agent constructor via the mcp_servers array.

Yes. You use the tts_sse tool for Server-Sent Events. This streams the generated speech chunks back to your application before the full sentence finishes processing.

It does. The stt_batch tool accepts audio files and returns text transcripts. Your OpenAI agent handles the resulting text for downstream analysis or summarization.

Call the get_usage_credits tool. Your agent pulls the exact billing stats. Set cacheToolsList=True in your setup so the agent knows this tool exists without pinging the server first.

The V8 Isolate Sandbox processes your five-second audio samples and generated speech bytes in pure memory. Vinkius destroys the ephemeral environment the moment your agent finishes the execution. Your raw audio never touches persistent disk storage.

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