Skip to content

Repository files navigation

Amadeus Logo

Amadeus AI

An interactive, voice-enabled AI companion recreating the digital memories of Makise Kurisu (Steins;Gate).


Amadeus Screenshot

🎥 Video Demo

Watch the Demo on YouTube

🌟 Overview

Amadeus AI is a modern web application built to bring the beloved "Amadeus" concept from Steins;Gate 0 to life. You can interact with a 3D VRM model of Makise Kurisu using your voice. She will listen, process your input with her distinct persona, and reply to you with synthesized voice and real-time lip-sync.

✨ Features

  • Makise Kurisu Persona: Powered by a highly customized Prompt using the Gemini 2.5 Flash API, ensuring her responses are analytical, intelligent, slightly tsundere, and completely in character.
  • Voice Interaction: Speak to her natively using the browser's free Web Speech API (Speech-to-Text).
  • Realistic Voice Synthesis: Uses the ElevenLabs API for high-quality, expressive text-to-speech.
  • 3D VRM Integration: Real-time rendering of Kurisu using Three.js with accurate lip-syncing mapped directly to the audio output.
  • Glassmorphism UI: A stunning, immersive, and collapsible chat interface.
  • Bilingual Support: Toggle between Thai and Japanese conversations seamlessly.

🚀 Technology Stack

  • Frontend: React, Vite
  • 3D Rendering: Three.js, @pixiv/three-vrm
  • LLM Engine: Google Gemini API (Gemini 2.5 Flash)
  • TTS Engine: ElevenLabs REST API
  • ASR Engine: Webkit Speech Recognition

⚙️ Installation & Setup

1. Prerequisites

2. Clone the Repository

git clone https://github.com/HakusaiTH/Amadeus.git
cd Amadeus

3. Install Dependencies

npm install

4. Configure Environment Variables

Create a .env file in the root directory and add your API keys:

VITE_GEMINI_API_KEY="your_gemini_api_key_here"
VITE_ELEVENLABS_API_KEY="your_elevenlabs_api_key_here"

(Note: Do not commit your .env file to version control. It is already added to .gitignore)

5. Run the Development Server

npm run dev

Open your browser and navigate to the local server address (usually http://localhost:5173).

🎮 How to Use

  1. Launch the App: The application will boot directly into the 3D scene.
  2. Select Language: Click the Chat icon at the bottom right to open the sidebar. Choose your preferred language (Thai or Japanese) from the top dropdown.
  3. Talk to Amadeus:
    • Voice: Click the floating Microphone Icon at the bottom left. Wait for it to turn red, then speak. It will automatically stop recording when you pause and send your message.
    • Text: Open the chat sidebar and type your message manually.
  4. Enjoy: Amadeus will ponder your input and reply with audio and lip-sync animations!

El Psy Kongroo.

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages