skills.rest
akostibas
akostibasCOMMUNITY
Explore all skills by this author →
GitHub

voice-mode

Transcribes live microphone input into complete utterances for Claude Code via Monitor tool without polling.

Updated Jun 10, 2026
Search Tags:#hands-free#claude-code#monitor#daemon#whisper-cpp#voice-mode#utterances
One-click install
npx skills add https://github.com/akostibas/voice-mode-skill --skill voice-mode-akostibas
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: voice-mode
Source: https://github.com/akostibas/voice-mode-skill/tree/main/skill
Command: npx skills add https://github.com/akostibas/voice-mode-skill --skill voice-mode-akostibas

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Hands-free voice input for Claude Code lets users start a local whisper.cpp daemon that transcribes microphone speech into complete utterances and delivers them to Claude Code via the Monitor tool, removing the need to type.

Core Features & Use Cases

  • Runs a background daemon that continuously transcribes microphone input into complete utterances.
  • Uses Monitor to tail a log and deliver each utterance to Claude Code as a chat notification.
  • Suppresses intermediate noise annotations and uses a deterministic gate to ensure only final utterances are delivered; supports pluggable backends and straightforward start/stop controls.

Quick Start

Start voice mode by saying "start voice mode" to launch the daemon, arm Monitor, and begin capturing complete utterances.

Frequently Asked Questions about voice-mode

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get hands-free voice input working with Claude Code?▼

Hands-free voice input for Claude Code is enabled by starting a local whisper.cpp daemon that transcribes microphone speech into complete utterances and delivers them directly via the Monitor tool. You can start it by saying "start voice mode".

How does whisper.cpp transcribe live microphone audio into text for coding sessions?▼

Whisper.cpp runs as a background daemon that continuously transcribes live microphone input into text. It applies a deterministic gate to suppress intermediate noise, ensuring only complete utterances are delivered as chat notifications.

Can I use voice mode for hands-free coding without polling for transcript updates?▼

Yes, voice mode uses the Monitor tool to tail a log file and deliver each transcribed utterance to Claude Code as a real-time chat notification, removing the need to manually poll for text updates.

Do I need to install whisper.cpp separately to use voice mode?▼

Voice mode requires a local whisper.cpp backend to run the transcription daemon. The skill includes an explicit update mechanism to upgrade the binary and skill, ensuring the background daemon operates with the correct dependencies.

How do I start, stop, or check the status of the voice transcription daemon?▼

Voice mode provides a CLI for straightforward start, stop, and status controls. You can say "start voice mode" to launch the daemon and arm Monitor, or use the CLI commands to manage the background transcription process.

Why does voice mode suppress intermediate noise annotations during transcription?▼

Voice mode suppresses intermediate noise annotations using a deterministic gate to ensure only final, complete utterances are delivered to Claude Code. This prevents partial speech fragments from interrupting your hands-free workflow.

Related Skills

Explore More Software Engineering Skills→
aviz85aviz85
45

live-transcribe

Stream microphone audio to ElevenLabs Scribe v2 Realtime and write transcribed text to a temporary file.

Community
Intermediate
smallnestsmallnest
601

coding-agent

Control multiple AI coding assistants via shell commands and PTY.

Community
Advanced
hafiznaveedchuhan-ctrlhafiznaveedchuhan-ctrl

streaming-llm-responses

Coordinate chat UI streaming events and thread lifecycle synchronization for LLM responses.

Community
Advanced
steipetesteipete
385.7k

openai-whisper

Transcribe local audio files to text or subtitles using the Whisper CLI.

Community
Intermediate