voicemode
VoiceMode lets you hold natural voice conversations with Claude Code and other MCP-capable agents. The README frames voice as being available when typing is not, such as walking to a meeting, cooking while debugging or giving your eyes a break. It needs a computer with a microphone and speakers.
There are two ways to install. The recommended one for Claude Code users is the plugin: `claude plugin marketplace add mbailey/voicemode`, `claude plugin install voicemode@voicemode`, then `/voicemode:install` and `/voicemode:converse`. The other is the Python installer, `uvx voice-mode-install`, followed by `claude mcp add --scope user voicemode -- uvx --refresh voice-mode`. Platforms listed are Linux, macOS, Windows (WSL) and NixOS, with Python 3.10 to 3.14, and system packages such as ffmpeg and portaudio are needed.
Features listed are low latency, smart silence detection that stops recording when you stop speaking, and a choice of privacy modes. Speech can run fully locally with Whisper.cpp for speech-to-text and Kokoro for text-to-speech, which expose the same API as OpenAI and switch seamlessly with it, or use OpenAI cloud services with `OPENAI_API_KEY`. The code is MIT licensed and was last pushed on 2026-10-06. This review is based on the README and repository data; we have not run it ourselves.
Features
Why choose it
Trade-offs
Maintenance check
- Last commit
- 2026-10-04
- Commits, last 90 days
- 100+
- Releases, last 90 days
- 1
- Latest release
- v8.12.0 (2026-07-21)
- Open issues
- 65
- Recent issues with at least one comment
- 8/21
Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.
Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.
Most discussed open issues
- #545 Windows MCP: recording never stops because lazy scipy.signal import stalls the VAD loop for ~24s4 comments · 2026-09-22
- #342 Mild crackling on WSLg after extended use4 comments · 2026-04-16
- #312 Support for push-to-talk and/or interruptible converse mode4 comments · 2026-03-11
- #211 Feature Request: Barge-in / Interrupt TTS playback when user starts speaking4 comments · 2026-01-22
- #511 VoiceMode Connect: POST /consent returns empty-bodied 503, MCP OAuth can never complete3 comments · 2026-08-04
Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.
Compatibility
Quick start
Use cases
Alternatives
Comments
No comments yet. Be the first!