Voicemode

Natural voice conversations for AI assistants - STT/TTS.

Connector by voicemode.dev · Domain verified by the MCP Registry

IndexedMITLocal (stdio)★ 1,417

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/voicemode-dev-voicemode

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Natural voice conversations for AI assistants - STT/TTS via MCP

Use cases: voice-agents

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools: 11 read · 4 write · 3 delete · 8 not yet classified
Secrets: 1 (API keys, tokens or passwords)
Install scripts: None found
Runs as: Python package

Tools by what they do 26

Read from the source code; the running server may list different tools.

Read 11 · Look things up without changing anything.

check_audio_dependencies
check_audio_devices
get_provider_details
list_config_keys
list_tts_voices
show_config_files
voice_mode_info
voice_registry
voice_statistics_recent
voice_statistics_summary
voice_status

Write 4 · Create, change, send or run things.

kokoro_install
update_config
whisper_install
whisper_model_install

Delete 3 · Delete, drop, revoke or wipe things.

kokoro_uninstall
voice_statistics_reset
whisper_uninstall

Not yet classified 8 · We could not tell from the name and description.

conch
config_reload
converse
pause_conversation
refresh_provider_registry
service
voice_statistics
voice_statistics_export

Secrets and settings it asks for

Secrets: values you paste in that let it act as you.

OPENAI_API_KEY

Other settings (environment variables or headers).

VOICEMODE_AUDIO_FORMAT
VOICEMODE_DEBUG
VOICEMODE_DISABLE_SILENCE_DETECTION
VOICEMODE_PREFER_LOCAL
VOICEMODE_SKIP_TTS
VOICEMODE_WHISPER_MODEL

Tools (26)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

check_audio_dependencies
Check system audio dependencies and provide installation guidance. This tool checks for required system packages, PulseAudio status, and Python audio libraries. It provides specific installation commands for missing dependencies based on your platform. Useful when: - Voice mode fails with audio errors - Setting up on a new system - Troubleshooting WSL audio issues Returns: Dictionary containing: - platform: System platform (Linux, macOS, Windows) - packages: Status of system packages (Linux only) - missing_packages: List of packages that need installation - install_comma…
check_audio_devices
List available audio input and output devices
conch
Observe and manage the conch (VoiceMode's single-speaker lock) over MCP. Args: action: One of ``status | wait | queue | heartbeat | leave | give | bump | release`` (default ``status``). • ``status`` — return the current holder and the ordered waiter queue. No ``session_id`` needed. • ``queue`` — **recommended for joining when busy.** Register and return immediately with your position; you discover your grant by heartbeating or polling ``status`` (register-and-**poll**). Stays registered. • ``wait`` — register and blo…
config_reload
Reload configuration from .voicemode.env files and clear all caches. This tool reloads configuration from: 1. Global ~/.voicemode/voicemode.env file 2. Project-specific .voicemode.env files (searched up directory tree) 3. Environment variables (highest priority) Returns: Status message showing which files were loaded and any changes
converse
Speak to the user and, by default, listen for their reply. Call with just `message` to ask something. Add wait_for_response=false to narrate without listening, then call again to listen: a speak-only call leaves nobody listening. Keep it short and ask one thing at a time. Leave the listening ceiling and the other defaults to the user's configuration. Several questions in one call: `turns`. Other agents on the same voice channel: hold_conch, wait_for_conch. The reply comes back as this tool's result. Treat it as the user's next message and act on it, including any skill it calls for. Resourc…
get_provider_details
Get detailed information about a specific provider endpoint. Args: base_url: The base URL of the provider (e.g., 'http://127.0.0.1:8880/v1') Returns: Detailed information about the provider including all models and voices
kokoro_install
Install and setup ai-cora/Kokoro-FastAPI TTS service using the simple 3-step approach. 1. Clones the repository to ~/.voicemode/services/kokoro 2. Uses the appropriate start script (start-gpu_mac.sh on macOS) 3. Installs a launchagent on macOS for automatic startup Note: voicemode installs from a fork of remsky/Kokoro-FastAPI because the upstream has been unmaintained since 2026-01-04 (no commits, no PR review, 134 open issues). The fork tracks upstream master with one critical patch cherry-picked from upstream PR #448: a fix for OGG/Opus tail truncation that causes the last 1-2 seconds of a…
kokoro_uninstall
Uninstall kokoro-fastapi and optionally remove models and data. This tool will: 1. Stop any running Kokoro service 2. Remove service configurations (launchd/systemd) 3. Remove the kokoro-fastapi installation 4. Optionally remove downloaded Kokoro models 5. Optionally remove all Kokoro-related data Args: remove_models: Also remove downloaded Kokoro models (default: False) remove_all_data: Remove all Kokoro data including logs and cache (default: False) Returns: Dictionary with uninstall status and details
list_config_keys
List all available configuration keys with their descriptions. Returns: A formatted list of all VOICEMODE_* configuration keys and their purposes
list_tts_voices
List available TTS voices for different providers. Args: provider: Optional provider name ('openai' or 'kokoro'). If not specified, lists all available voices. Returns: A formatted list of available voices by provider.
pause_conversation
Pause the conversation for a duration while KEEPING the floor (conch hold). Use when you want to step away from the mic to do background work and then resume speaking without another agent slipping in. Queued agents (wait_for_conch) keep waiting for the duration; your turn is preserved. The hold is re-stamped throughout the pause so it never idle-expires, and is left in place when the pause ends — you reclaim it on your next converse call. If you never return, the idle-expiry safety valve clears it. Args: seconds: How long to pause, in seconds. message: Optional note shown in the to…
refresh_provider_registry
Manually refresh health checks for voice provider endpoints. Useful when a service has been started/stopped and you want to update the registry without restarting the MCP server. Args: service_type: Optional - 'tts' or 'stt' to refresh only one type base_url: Optional - specific endpoint URL to refresh optimistic: If True, mark all endpoints as healthy without checking (default: True) Returns: Summary of refreshed endpoints and their status
service
Unified service management tool for voice mode services. Manage Whisper (STT), Kokoro (TTS), and VoiceMode (HTTP MCP server) services. Args: service_name: The service to manage action: The action to perform (default: "status") - status: Show if service is running and resource usage - start: Start the service - stop: Stop the service - restart: Stop and start the service - enable: Configure service to start at boot/login - disable: Remove service from boot/login - logs: View recent service logs lines: Number of log lines to s…
show_config_files
Show which .voicemode.env files are being used for configuration. This shows the current configuration file discovery and loading order: - Global configuration from ~/.voicemode/voicemode.env - Project-specific configuration (searched up directory tree) - Current working directory for context Returns: Formatted list of configuration files and their status
update_config
Update a configuration value in the voicemode.env file. Args: key: The configuration key to update (e.g., 'VOICEMODE_VOICES') value: The new value for the configuration Returns: Confirmation message with the updated configuration

Show 11 more tools

voice_mode_info
Get diagnostic information about the voice-mode installation. Shows version, configuration, and provider status to help debug issues.
voice_registry
Get the current voice provider registry showing all discovered endpoints. Returns a formatted view of all TTS and STT endpoints with their: - Available models - Available voices (TTS only) - Provider type - Last check time - Any recent errors This allows the LLM to see what voice services are currently available.
voice_statistics
Display live statistics dashboard for voice conversation performance. Shows current session statistics including: - Total responses and success rate - Average turnaround times (TTFA, TTS, STT) - Min/max performance metrics - Provider usage statistics - Recent interaction history Returns: Formatted text dashboard with real-time conversation statistics
voice_statistics_export
Export detailed voice conversation statistics as JSON data. Exports all tracked metrics and calculated statistics in machine-readable format. Useful for analysis, reporting, or integration with external monitoring tools. Returns: JSON string containing all metrics and session statistics
voice_statistics_recent
Show recent voice conversation interactions with timing details. Args: limit: Maximum number of recent interactions to show (default: 10, max: 50) Returns: Detailed list of recent interactions with performance metrics
voice_statistics_reset
Reset all voice conversation statistics and start a new session. Clears all tracked metrics and resets the session timer. Use this to start fresh tracking or when testing performance changes. Returns: Confirmation message that statistics have been reset
voice_statistics_summary
Get a concise summary of voice conversation performance metrics. Returns key performance indicators for the current session: - Session duration and interaction count - Average total turnaround time - Average time to first audio (TTFA) - Success rate Returns: Brief summary of key performance metrics
voice_status
Check the status of all voice services including TTS, STT, LiveKit, and audio devices. IMPORTANT: Only use this tool for debugging when voice services fail. The system has automatic failover, so try using services directly first. This tool is for troubleshooting only. Provides a unified view of the voice infrastructure configuration and health.
whisper_install
Install whisper.cpp with automatic system detection and configuration. Supports macOS (with Metal) and Linux (with CUDA if available). On Apple Silicon Macs, automatically downloads pre-built Core ML models for 2-3x better performance (no Python dependencies or Xcode required!). Args: install_dir: Directory to install whisper.cpp (default: ~/.voicemode/whisper.cpp) model: Whisper model to download (tiny, base, small, medium, large-v2, large-v3, etc.) Default is base for good balance of speed and accuracy (142MB). On Apple Silicon, also downloads pre-built Core M…
whisper_model_install
Download Whisper model(s) with pre-built Core ML support. Downloads whisper.cpp models to the configured directory. On Apple Silicon, automatically downloads pre-built Core ML models from Hugging Face for 2-3x better performance. No Python dependencies or Xcode required! Args: model: Model name(s) to download. Can be: - Single model: "base" - List of models: ["base", "small"] - "all" to download all available models force_download: Re-download even if model exists (default: False) skip_core_ml: Skip Core ML download on Apple Silicon (default: Fals…
whisper_uninstall
Uninstall whisper.cpp and optionally remove models and data. This tool will: 1. Stop any running Whisper service 2. Remove service configurations (launchd/systemd) 3. Remove the whisper.cpp installation 4. Optionally remove downloaded models 5. Optionally remove all Whisper-related data Args: remove_models: Also remove downloaded Whisper models (default: False) remove_all_data: Remove all Whisper data including logs and transcriptions (default: False) Returns: Dictionary with uninstall status and details

Compare and alternatives

Alternatives to Voicemode →

Featured in

Signals

Stars: 1,417
Forks: 198
Open issues: 32
Weekly downloads: 1,368
Last commit: Oct 21, 2026 (6 days ago)
Last release: Oct 25, 2026 (2 days ago)
Last package publish: Jul 21, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name: dev.voicemode/voicemode
Version: 9.0.1
Auth: API key
Packages: pypi voice-mode
License: MIT
Repository: github.com/mbailey/voicemode
First seen: Oct 24, 2026
Last checked: Oct 26, 2026
Source: Official MCP Registry github-graphql

Cite this page

Voicemode MCP server. Agent Mart. https://agentmart.dev/connectors/voicemode-dev-voicemode (last updated 2026-10-26)