Skip to content

Releases: kenta4es/voice-claude-kenta4es

Release list

v1.0.0 — Voice I/O for Claude Desktop

v1.0.0 — Voice I/O for Claude Desktop Latest
Latest

Choose a tag to compare

View all tags
@kenta4es kenta4es released this 19 May 18:10
v1.0.0
e3ff9e1
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Initial public release.

Highlights

  • Auto-speak in Claude Desktop — every Claude reply is read aloud via the voice-output skill and claude-tts MCP. Russian neural voices out of the box (Dmitry, Svetlana, Irina).
  • AutoHotkey hotkeys — LCtrl+LAlt+Z/A/C/↑↓/←→ for pause, stop, mute, rate, voice cycling. Layout-independent (RU/EN). Work in any window.
  • Independent TTS server — autostarts at logon via Scheduled Task, lives outside Claude Desktop. AHK hotkeys keep working when Claude is closed.
  • HTTP API on 127.0.0.1:48329 — any local app can drive TTS (PowerShell, Python, browser bookmarklets).
  • Single-instance HTTP delegation architecture — solves the "MCP speaks but hotkeys don't pause" mismatch by routing both paths to one engine.
  • Optional voice dictation — OpenWhispr + Groq Whisper Large v3 Turbo (free tier). Setup guide in docs/.

Install

  • Cowork plugin users: download voice-claude-kenta4es.plugin below, drag into Claude Desktop, run post-install script.
  • Standalone: clone the repo, run standalone\install.ps1.

See README.md for full instructions, hotkeys, troubleshooting.

Authors

Kenta4es & Claude (Anthropic), built together in one Cowork session.

Assets 3
Loading