Skip to main content
Glama
kubouchiyuya

claude-voice-notify

by kubouchiyuya
OverviewSchemaRelated ServersScoreDiscussions
Local

Typing SVG

GitHub X LP

Stars Forks License

What This Does

claude-voice-notify turns Claude Code into a voice-aware workflow. It speaks when a task completes, when a confirmation is needed, or when an error appears, so you can keep moving even if you are away from the keyboard.

The default engine is VOICEVOX, which runs locally and costs nothing to use. If you want to change the sound or the provider, the repository supports eight backends and a guided setup flow for non-engineers.

Related MCP server: Voicevox MCP Server

Why This Exists

MASA builds tools from the perspective of a salon owner who uses AI as the execution layer.

This repository is one of the interfaces in that system:

Layer

Role

Human intent

MASA defines the outcome

Claude Code

Implements the work

Voice layer

Tells the human what happened

Provider layer

Selects the best TTS engine

✨ Features

Feature

Description

8 providers

VOICEVOX, COEIROINK, AivisSpeech, Style-Bert-VITS2, Google TTS, Fish Audio, ElevenLabs, OpenAI

Local default

VOICEVOX runs locally with no API key

Star gate

GitHub Star verification for community access control

Language switch

Japanese and English notification output

Terminology mode

Explains technical terms in plain language

VS Code support

Toggle notification behavior from the editor

One-command update

Keep the setup current with ./update.sh

🚀 Quick Start

Star First

This tool is designed with a GitHub Star gate. Star the repo before setup.

Setup

cd ~/Projects
git clone https://github.com/kubouchiyuya/claude-voice-notify.git
cd claude-voice-notify
./setup.sh

Test

./notify.sh "Setup successful"

Hooks

Add this to ~/.claude/settings.json:

{
  "hooks": {
    "Stop": [
      {
        "matcher": "",
        "hooks": [
          {
            "type": "command",
            "command": "/path/to/claude-voice-notify/notify.sh 'Task complete' complete"
          }
        ]
      }
    ],
    "Notification": [
      {
        "matcher": "",
        "hooks": [
          {
            "type": "command",
            "command": "/path/to/claude-voice-notify/notify.sh 'Please confirm' confirm"
          }
        ]
      }
    ]
  }
}

🏗️ Architecture

graph TD
    A[Claude Code] -->|Hook: Stop / Notification| B[notify.sh]
    A -->|MCP tool call| C[MCP server dist/index.js]
    B --> D{Provider router}
    C --> D
    D -->|voicevox| E[VOICEVOX local API]
    D -->|coeiroink| F[COEIROINK local API]
    D -->|aivis_speech| G[AivisSpeech local API]
    D -->|style_bert_vits2| H[Style-Bert-VITS2]
    D -->|google| I[Google Cloud TTS]
    D -->|fish_audio| J[Fish Audio API]
    D -->|elevenlabs| K[ElevenLabs API]
    D -->|openai| L[OpenAI TTS API]
    E --> M[Playback: afplay / paplay / PowerShell]
    F --> M
    G --> M
    H --> M
    I --> M
    J --> M
    K --> M
    L --> M

🛠️ Tech Stack

TypeScript Node.js MCP VOICEVOX Shell

👤 Author

Yuya Kubouchi (MASA2 / MASA Sub)

Beauty Salon Owner → AI Architect

Location

Osaka, Japan

System

AKATSUKI / Miyabi Society

Focus

AI × Beauty Industry × Non-Engineer Operations

"I do not code first. I design the system, and AI executes it."

If this saved you time, please give it a star.

This server cannot be installed

A
license - permissive license
-
quality - not tested
C
maintenance

How are these scores calculated?

Maintenance

–Maintainers
–Response time
–Release cycle
–Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/kubouchiyuya/claude-voice-notify'

If you have feedback or need assistance with the MCP directory API, please join our Discord server