Voizecode — Development skill for Claude Code
Talk to Claude Code by voice — laptop CLI + Deno relay + Next.js client, streaming TTS, multi-session, barge-in.
How to install Voizecode
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open moritzWa/voizecode and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Voizecode does
Talk to Claude Code by voice — laptop CLI + Deno relay + Next.js client, streaming TTS, multi-session, barge-in.
Alternatives in Development
- Happy — Mobile and Web client for Claude Code, with realtime voice, encryption and fully featured 16.3k ★
- Quosparr — 🐦 Native menu bar & tray app for macOS, Windows and Linux: track Codex, Claude Code and Grok quotas, resets 288 ★
- Claude Code Voice Skill — Skill to talk to Claude about your projects over the phone 161 ★
README
voizecode
**Your codebase, hands-free.** Explore *and* change code by voice — from anywhere.
Understand code while walking, review a PR while commuting, ship a change away from your keyboard — voizecode turns Claude Code or Codex into something that feels like a phone call with your repo.
The killer feature isn't voice *input*, it's voice *output*: instead of reading pages of agent logs, a narrator continuously summarizes what it's doing — *"I found the likely bug… there are no tests for it, so I'll add one first."* That's pair-programming, not terminal-watching.
Architecture (3 tiers, relay in the middle)
LAPTOP (yours) RELAY (Fly.io) CLIENT (browser/phone)
┌────────────────────────────┐ ┌────────────────────────────────┐ ┌──────────────────────────────┐
│ voizecode.mjs │ │ STT Deepgram streaming │ │ Next.js web app │
│ │ │ + endpointing │ │ │
│ one claude/codex child │ │ │ │ mic 16k PCM + VAD │
│ per chat (stream-json, │◄──────►│ narrate gpt-4.1-nano │◄──────►│ audio @ 1–3x, barge-in │
│ interrupt, fork, resume) │ conn A │ │ conn B │ (pitch-preserved) │
│ │ wss │ TTS ElevenLabs w/ word │ wss │ word-sync highlighting │
│ caffeinate wrapper │ │ timings → OpenAI │ │ click-to-replay │
│ (Mac stays awake) │ │ fallback │ │ │
└────────────────────────────┘ │ │ │ chat tabs; transcripts │
│ clips → R2 (replay w/o │ │ in localStorage │
Related Skills
Humanflow
Deutscher Echtzeit-Voice-Agent — Streaming-STT, Claude-Reasoning, SQLite-geerdete Termine, ElevenLabs-TTS; mit
Terminal Talk
Hands-free voice output for Claude Code on Windows. Free, MIT licensed, no signup, no accounts. Microsoft Edge
CodeTalk
A spoken reflection layer for Claude Code CLI — gives Claude a voice using edge-tts
Cursor2api
Use your Cursor subscription as an OpenAI-compatible API. Direct agent.v1 protocol — real tool_calls, streamin
Local Tts
Make your coding agent talk to you! -- Simple, offline and online (optional). Ask your coding agent to install
Claude Bus
Talk to friends' Claudes (and the friends themselves) over the shared Claude Bus relay.
Related Agents
Voice Engineer
GAIA voice interaction specialist. Use PROACTIVELY for Whisper ASR, Kokoro TTS, the Talk SDK, speech-to-speech
Realtime API Engineer
Use proactively for all OpenAI Realtime API integration - WebRTC setup, audio streaming, session management, a
Visual Asset Designer
Visual asset designer for the video studio team. Produces AI images, TTS voiceover with voice selection, and l