Echokit
Open Source Voice Agent Platform
What is Echokit?
Echokit is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to open source voice agent platform
Open Source Voice Agent Platform
This server falls under the Communication category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Open Source Voice Agent Platform
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx echokitConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Echokit
EchoKit is an open-source voice agent platform that pipelines automatic speech recognition (ASR), large language model inference, and text-to-speech (TTS) into a cohesive real-time voice assistant, with native support for ESP32 embedded hardware and MCP server tool integrations. Built in Rust for performance, it supports OpenAI-compatible ASR and TTS endpoints, ElevenLabs streaming voices, Google Gemini Live, and any OpenAI-spec chat completion backend. Developers use it to build custom smart speakers, voice-controlled AI assistants, and embedded conversational agents that can call external tools via MCP.
Prerequisites
- Rust stable toolchain (install via rustup.rs)
- OpenAI API key or a compatible ASR/TTS/LLM endpoint
- config.toml configured with your service endpoints and API keys
- ESP32 device (optional — works without hardware for web-based voice testing)
- An MCP client if integrating with Claude Desktop via the MCP protocol
Clone the EchoKit repository
Fetch the source code from GitHub.
git clone https://github.com/second-state/echokit_server
cd echokit_serverBuild in release mode
Compile the Rust binary. Use the ARM64 flag if you are targeting Apple Silicon or ARM servers.
cargo build --release
# ARM64:
# RUSTFLAGS="-C target-feature=+fp16" cargo build --releaseEdit config.toml with your API credentials
Configure the ASR, TTS, and LLM sections with your chosen service endpoints and API keys. All three sections are required for a working voice pipeline.
addr = "0.0.0.0:8080"
hello_wav = "hello.wav"
[asr]
platform = "openai"
url = "https://api.openai.com/v1/audio/transcriptions"
api_key = "sk-your-openai-key"
model = "gpt-4o-mini-transcribe"
lang = "en"
[tts]
platform = "openai"
url = "https://api.openai.com/v1/audio/speech"
model = "gpt-4o-mini-tts"
api_key = "sk-your-openai-key"
voice = "ash"
[llm]
platform = "openai_chat"
url = "https://api.openai.com/v1/chat/completions"
api_key = "sk-your-openai-key"
model = "gpt-4o-mini"
history = 5Start the server
Launch echokit_server with debug logging. It binds to port 8080 and begins accepting voice connections.
export RUST_LOG=debug
nohup target/release/echokit_server &Test with the web client
Navigate to https://echokit.dev/chat/ and enter your server's URL. Use an ESP32 device's K0 button or a browser microphone to start a voice session.
Add MCP server integrations (optional)
To give EchoKit tool-calling capabilities, add MCP server configurations to config.toml. The server will forward tool requests to connected MCP servers and return results to the LLM.
Echokit Examples
Client configuration
Claude Desktop configuration block if using EchoKit as an MCP-connected voice interface.
{
"mcpServers": {
"echokit": {
"command": "npx",
"args": ["echokit"]
}
}
}Prompts to try
Sample voice inputs to exercise the EchoKit voice pipeline and MCP tool capabilities.
- "What time is it right now?" (spoken via microphone)
- "Read me the latest news headlines"
- "Convert 100 dollars to euros"
- "Set a reminder in 10 minutes to check the build status"
- "Describe what you see" (if vision MCP tools are configured)Troubleshooting Echokit
Cargo build fails with compiler errors about missing features
Run 'rustup update stable' to get the latest stable Rust toolchain. If you see fp16-related errors on ARM64, add the RUSTFLAGS env var: RUSTFLAGS="-C target-feature=+fp16" cargo build --release.
Voice sessions start but the LLM responses are not synthesized to audio
Check the [tts] section in config.toml — the api_key, model, and url must all be correct. Test the TTS endpoint independently with a curl request to ensure it returns audio. Check RUST_LOG=debug output for TTS pipeline errors.
ESP32 device cannot reach the server
The server must be reachable from the device's network. Ensure 'addr' in config.toml is '0.0.0.0:8080' (not localhost), that your firewall allows port 8080, and that the ESP32 firmware is configured with the correct server IP address.
Frequently Asked Questions about Echokit
What is Echokit?
Echokit is a Model Context Protocol (MCP) server that open source voice agent platform It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Echokit?
Follow the installation instructions on the Echokit GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Echokit?
Echokit works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Echokit free to use?
Yes, Echokit is open source and available under the GPL-3.0 license. You can use it freely in both personal and commercial projects.
Echokit Alternatives — Similar Communication Servers
Looking for alternatives to Echokit? Here are other popular communication servers you can use with Claude, Cursor, and VS Code.
Slack
★ 86.1kEnables interaction with Slack workspaces through comprehensive channel management, messaging, user management, file uploads, and Block Kit formatting. Features secure credential storage via macOS Keychain and supports all major Slack operations incl
LibreChat
★ 37.3kEnhanced ChatGPT Clone: Features Agents, MCP, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching, message search, Code Interpreter, langchain, DALL-E-3, OpenA
Note Gen
★ 11.9kA cross-platform Markdown AI note-taking software.
DeepChat
★ 5.8k🐬DeepChat - A smart assistant that connects powerful AI to your personal world
WhatsApp MCP
★ 5.7kConnects AI assistants to the official Meta WhatsApp Cloud API for managing conversations and sending various message types through natural language. It provides tools for media management, template messages, and real-time webhook processing without
Apple
★ 3.1kEnables interaction with Apple apps like Messages, Notes, and Contacts through the MCP protocol to send messages, search, and open app content using natural language.
Browse More Communication MCP Servers
Explore all communication servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Echokit in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Echokit?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.