AI Waifu VRM (Alpha)
❗❗❗❗❗❗
This project is not open source and is licensed under All Rights Reserved.
You may not reproduce, copy, modify, distribute, or use the source code for commercial purposes without explicit permission from Mustafa.
The software is only intended for testing and use as provided via the original GitHub Pages
❗❗❗❗❗❗
GitHub Pages link: https://mustafaincby44.github.io/Ai-Waifu-VRM/
This project is an advanced chatbot application that can converse with a 3D anime character, powered by Edge TTS and Gemini AI. ✨ Features
3D Anime Character: 3D anime girl model in VRM format
Edge TTS: Microsoft's free, unlimited TTS service
Gemini AI: Intelligent conversation powered by Google's advanced AI model
Multilingual Support: Turkish, English, Japanese, Korean, Chinese, and more
Mouth Animation: Mouth movement with real-time voice analysis
Emotion System: Expressing the character's emotions
Brain System: Advanced brain system integration
API Tracking: Usage limits and tier management
🎵 TTS Models Edge TTS (Default)
Free and 50 daily requests (free model 50)
9 language support: Turkish, English, Japanese, Korean, German, French, Spanish, Italian, Russian
Male/female voice options for each language
Adjustable speech speed (0.5x - 2.0x)
Gemini TTS
15 requests per day limit (free-tier API 15)
High-quality AI voice
Various voice options
🌍 Supported Languages Language Code Voices (ngrok with local edge tts)
Turkish tr Emel (F), Ahmet (M)
English en Jenny (F), Guy (M), Ana (F), Aria (F), Christopher (M), Eric (M), Michelle (F), Roger (M), Natasha (F), William (M), Clara (F), Liam (M), Libby (F), Maisie (F), Ryan (M), Sonia (F), Thomas (M), Sam (F), Yan (M), Connor (M), Emily (F), Neerja (F), Prabhat (M), Asilia (F), Chilemba (M), Abeo (M), Ezinne (F), Mitchell (M), James (M), Rosa (F), Luna (F), Wayne (M), Elimu (M), Imani (F), Leah (F), Luke (M)
Japanese ja Nanami (F), Keita (M)
Korean ko Sun-Hi (F), InJoon (M)
German de Katja (F), Amala (F), Conrad (M), Killian (M), Ingrid (F), Jonas (M), Jan (M), Leni (F)
French fr Denise (F), Eloise (F), Henri (M), Sylvie (F), Antoine (M), Jean (M), Ariane (F), Fabrice (M), Charline (F), Gerard (M)
Spanish es Elena (F), Tomas (M), Marcelo (M), Sofia (F), Gonzalo (M), Salome (F), Juan (M), Maria (F), Belkys (F), Emilio (M), Ramona (F), Andrea (F), Luis (M), Alvaro (M), Elvira (F), Teresa (F), Andres (M), Marta (F), Carlos (M), Karla (F), Federico (M), Yolanda (F), Margarita (F), Roberto (M), Alex (M), Camila (F), Karina (F), Victor (M), Mario (M), Tania (F), Lorena (F), Rodrigo (M), Alonso (M), Paloma (F), Mateo (M), Valentina (F), Paola (F), Sebastian (M)
Italian it Isabella (F), Diego (M), Elsa (F)
Russian ru Svetlana (F), Dmitry (M)
⚙️ Settings TTS Model Selection
Click the Settings button
Select Edge TTS or Gemini TTS from the “TTS Model” option
If Edge TTS is selected:
Select language
Select voice
Adjust speech speed
API Keys
Response API Key: For Gemini AI
TTS API Key: For Gemini TTS (must be different)
🔧 Advanced Features
With Edge TTS, you can set the speech speed between 0.5x and 2.0x. Fallback System
If TTS fails, speech continues with text simulation.
📁 File Structure
AiProject/
├── app.js # Main JavaScript application
├── brain.js # Brain system
├── index.html # HTML interface
├── styles.css # CSS styles
├── tts_server.py # Edge TTS Python backend
├── requirements.txt # Python packages
└── README.md # This file
🌐 Edge TTS Server
The Edge TTS server is exposed to the outside world via ngrok:
Endpoint: /speak
Method: POST
Port: 5000 (local)
📄 License
This project is licensed under the “all rights reserved” license. See the LICENSE file for details.
🙏 Thanks
Microsoft Edge TTS - Free TTS service
Google Gemini AI - AI model
Three.js - 3D graphics library
VRM - 3D model format
📌 TODO / Future Improvements
Live microphone input support
Custom Character Import
better web interface
Anims