How to Use the Cartesia (Voice AI) MCP in Google ADK

Q: Can I filter which Cartesia (Voice AI) tools Google ADK sees?

Yes. The toolset accepts an optional toolnames filter. You restrict the agent to just ttsbytes and sttbatch if you want to block it from deleting voices.

Q: Does Cartesia (Voice AI) handle batch transcription in Google ADK?

It handles it natively. The sttbatch tool converts large audio files into text. Your Gemini agent reads the output and can immediately insert the transcript into BigQuery.

Q: How do I stitch audio segments together?

Use the infillbytes tool. It generates the missing speech between two existing audio files. The transition sounds natural because the model matches the surrounding acoustic environment.

Give your Google ADK agents native voice capabilities. Clone speech, transcribe calls, and generate audio inside your Google Cloud environment.

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 Google ADK

Create your Vinkius account to connect Cartesia (Voice AI) to Google ADK 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 Google ADK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Run the Cartesia MCP Server in Google ADK

Gemini models hold massive context windows. Now they can speak. You attach this MCP server to your LlmAgent, and it instantly gains access to `tts_bytes` for high-quality audio generation directly from your text inputs. The integration handles complex voice mapping. Your agent calls `list_voices` to grab available profiles, then uses `localize_voice` to adapt them across dialects. You feed BigQuery text data right into the audio generator without leaving your cloud perimeter.

Clone voices from five-second clips

Building custom voice profiles takes seconds, not hours. The `clone_voice` tool accepts a short audio snippet and creates a reusable voice identity. If you make a mistake, `delete_voice` removes it from your account immediately. You often need to modify existing recordings. The `voice_changer_bytes` tool swaps out the speaker in an audio file but leaves the original emotion and pacing alone. Gemini analyzes the transcript, and Cartesia handles the audio swap.

Manage enterprise pronunciations

Corporate acronyms ruin text-to-speech outputs. You fix this using `create_pronunciation_dict` to define custom phonetic rules. Your agent reads a list of terms from Vertex AI and builds the dictionary automatically. Auditing matters for enterprise voice deployments. The `list_agent_calls` tool pulls historical call data and transcripts for specific voice agents. You pipe those transcripts straight back into Gemini's massive context window for deep analysis.

Setup guide

Set up Cartesia (Voice AI) MCP in Google ADK

Prerequisites

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

1

Install Google ADK

Run pip install google-adk to install the Agent Development Kit. MCP support is included via the McpToolset class.
2

Connect via SSE transport

Use McpToolset.from_server() with SseServerParams pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create an LlmAgent

Pass the returned mcp_tools list directly to LlmAgent(tools=mcp_tools). The ADK maps each MCP tool to a native Gemini function call — no manual schema definitions required.
4

Run with any Gemini model

The agent works with any Gemini model (gemini-2.0-flash, gemini-2.5-pro, etc.). Copy the full example on the right to get started with Cartesia (Voice AI) tools in your ADK agent.

agent.py

from google.adk.agents import LlmAgent
from google.adk.tools.mcp_tool.mcp_toolset import McpToolset
from google.adk.tools.mcp_tool.mcp_session_manager import SseServerParams

# Connect to the MCP via SSE
mcp_tools, exit_stack = await McpToolset.from_server(
    connection_params=SseServerParams(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    )
)

# Create your agent with auto-discovered tools
agent = LlmAgent(
    name="Cartesia (Voice AI)_agent",
    model="gemini-2.0-flash",
    instruction="You have access to Cartesia (Voice AI) tools via MCP.",
    tools=mcp_tools,
)

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 Google ADK

Install google-adk via pip. Create an McpToolset using StreamableHttpServerParameters with your Vinkius URL. Assign that toolset to the tools parameter in your LlmAgent setup.

Yes. The toolset accepts an optional tool_names filter. You restrict the agent to just tts_bytes and stt_batch if you want to block it from deleting voices.

It handles it natively. The stt_batch tool converts large audio files into text. Your Gemini agent reads the output and can immediately insert the transcript into BigQuery.

Use the infill_bytes tool. It generates the missing speech between two existing audio files. The transition sounds natural because the model matches the surrounding acoustic environment.

Vinkius routes your five-second voice samples through a zero-trust architecture. The managed endpoint authenticates your single token, processes the clone request, and immediately drops the connection. No audio data persists in the middle layer.

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