Four interactive demos showcasing the core capabilities of the MascotBot SDK for React.
- Generated Voice-Over — play pre-generated audio with synchronized lip-sync visemes
- Input Voice-Over — type text and generate speech with real-time lip-sync via TTS API
- Voice Call with AI — real-time voice conversation with an AI-powered mascot (LLM + TTS)
- Rive Input Controls — directly control Rive state machine inputs (boolean, numeric, trigger) and dynamically load assets
- Node.js 18+
- pnpm (or npm/yarn)
- Mascot Bot SDK subscription (for
.tgzpackage) - Rive
.rivanimation files (for input-voice-over and rive-inputs demos) - ElevenLabs API key (for voice call demo)
- Groq or OpenAI API key (for voice call demo)
- Clone this repository
- Add the SDK package and Rive files (see below)
- Configure environment variables
- Install and run
pnpm install
pnpm devOpen http://localhost:3000 to see all four demos.
- File:
mascotbot-sdk-react-X.X.X.tgz - Where: project root
- How to get: download from your Mascot Bot dashboard after subscribing
cp /path/to/mascotbot-sdk-react-X.X.X.tgz ./
pnpm install- Files:
bear.riv,notionGuys.riv,mascot-cat-customizible.riv(and others) - Where:
public/demo_mascots/ - How to get: provided with your Mascot Bot SDK subscription
- Note: the generated voice-over demo loads its Rive file from a CDN, so no local file is needed for that demo
cp /path/to/*.riv ./public/demo_mascots/Copy .env.example to .env.local (CRA uses .env.local or .env):
cp .env.example .env.local| Variable | Description | Required |
|---|---|---|
REACT_APP_MASCOTBOT_API_KEY |
Mascot Bot API key | Yes (for input voice-over and call demos) |
REACT_APP_ELEVENLABS_API_KEY |
ElevenLabs TTS API key | For voice call demo |
REACT_APP_LLM_API_KEY |
Groq or OpenAI API key | For voice call demo |
REACT_APP_SPREADSHEET_ID |
Google Sheets ID for conversation history | Optional |
Uses pre-recorded audio + visemes JSON to demonstrate lip-sync playback. No API keys needed.
Type text, send to MascotBot TTS API, and watch the bear mascot speak with synchronized lip movements.
Full real-time voice conversation: microphone input -> LLM -> TTS -> lip-synced mascot. Requires all API keys.
Directly manipulate Rive state machine inputs: toggle boolean (violet eyes), adjust numeric (headphones color), fire triggers (thumbs up), and dynamically swap image assets.
See LICENSE.