Skip to content

Quick Start

Prerequisites

To use AITuberKit, you need the following software:

  • Node.js: ^20.0.0
  • npm: ^10.0.0

Installation Steps

1. Clone the Repository

bash
git clone https://github.com/tegnike/aituber-kit.git
cd aituber-kit

2. Install Packages

bash
npm install

3. Set Environment Variables (Optional)

If needed, copy the .env.example file to .env and set the environment variables.

bash
cp .env.example .env

INFO

Values set in environment variables have lower priority than values entered in the settings screen.

Start the Development Server

bash
npm run dev

You can start using AITuberKit by opening http://localhost:3000 in your browser.

Screen Description

Screen Description

  • The settings button with a gear icon at the top left allows you to configure various settings related to AI characters and voice
  • The conversation log button (right of the settings button) allows you to display the conversation history with the character by clicking it (adjustable width by dragging)
  • The character is displayed in the center, and its size and position can be freely adjusted by dragging
  • There is an input form with a microphone icon at the bottom, which allows you to text or voice input to the character
  • Click the send button (right arrow) or press Enter to send a message

Basic Usage

Interacting with AI Characters

  1. Enter the API key for the LLM selected in the settings screen
  2. Edit the character's setting prompt as needed
  3. Upload a VRM file or Live2D file for the character, and a background file if needed
  4. Select a voice synthesis engine and configure voice settings as needed
  5. Start a conversation with the character from the input form

AITuber Streaming

  1. Turn ON Youtube mode in the settings screen
  2. Enter your Youtube API key and Youtube Live ID
  3. Configure other settings similar to "Interacting with AI Characters"
  4. Start streaming on Youtube and confirm that the character responds to comments
  5. Turn ON conversation continuation mode to allow the AI to speak proactively when there are no comments

Next Steps