Welcome to Concept Café Simulator
Concept Café Simulator is an AI talk game where you create concept café staff members, known as “casts,” and enjoy conversations with them.
Because the AI responds as the cast, every conversation is different.
Cast Creation
Choose from five base avatars and customize details such as hairstyle, clothing, and accessories. The game also supports importing VRM files, allowing you to use your favorite 3D models as casts.
Create each cast’s personality by setting their name, first-person pronoun, and speaking style. You can choose from 10 speaking styles, including “Polite,” “Casual,” “Elegant Lady,” “Tsundere,” and “Yandere.” You can also set their favorite topics from over 100 categories, covering a wide range of genres such as anime, games, music, cooking, and sports.
Conversations with AI
The game includes a local LLM, allowing you to have conversations without an internet connection. Your conversation data is never sent to external servers. In addition to text input, voice recognition input is also supported.
Casts Remember You
Casts learn and remember information from your conversations with them. They can remember things you have talked about, such as your favorite foods, hobbies, and work, and use that information in future conversations.
Depending on the number of visits and the content of your conversations, 12 emotional parameters such as affection, trust, interest, and excitement will change, deepening your relationship with each cast.
Voice Synthesis
Japanese voice synthesis powered by VOICEVOX is included. You can choose from 15 different voices for each cast. Cast expressions change automatically based on the conversation, and lip-sync is also supported.
Five Cafés
The game includes five café spaces, each with a different theme. You can assign a different café to each cast.
Main Features
AI-powered real-time conversations
Local processing and offline support
Five base avatars plus VRM import support
Conversation memory and relationship system
VOICEVOX voice synthesis with 15 voices
Microphone input via voice recognition
Emotion-linked facial expressions and lip-sync
Five themed cafés
