converse
Verified Safeby fwdslsh
Overview
A lightweight Model Context Protocol (MCP) server that provides voice capabilities using remote OpenAI-compatible TTS/STT APIs, enabling AI assistants like Claude to speak and listen through the system's audio devices.
Installation
converseEnvironment Variables
- STT_API_URL
- TTS_API_URL
Security Notes
The server uses `child_process.spawn` to execute external system commands (`aplay`, `arecord` by default) for audio playback and recording. The commands themselves, `RECORD_CMD` and `PLAY_CMD`, are fully configurable via environment variables. While `spawn` with an array of arguments is generally safer against shell injection from *tool parameters*, a malicious actor with control over these environment variables could configure them to execute arbitrary commands on the host system, leading to a significant arbitrary code execution vulnerability. There are no obvious hardcoded secrets, and API keys are read from environment variables. No `eval` or code obfuscation was detected.
Similar Servers
elevenlabs-mcp
An official ElevenLabs Model Context Protocol (MCP) server that enables AI clients (e.g., Claude Desktop, Cursor) to interact with ElevenLabs Text-to-Speech, voice cloning, transcription, and audio processing APIs.
voicemode
A standalone installer package for VoiceMode that handles system dependency detection and installation, and provides voice interaction capabilities (STT/TTS) for Model Context Protocol (MCP) servers.
deepl-mcp-server
Provides translation and rephrasing capabilities using the DeepL API, integrated as a Model Context Protocol (MCP) server for AI clients.
mcp-tts
Provides a Text-to-Speech (TTS) server via the Model Context Protocol (MCP) to integrate speech synthesis into AI agents like Claude Desktop and Cursor IDE.