Living Memories - Inworld Node Runtime Demo

Create talking-photo experiences powered by Inworld's Voice Clone, TTS, and Node.js Runtime, combined with video generation from Runway ML. This template demonstrates two key features: an interactive Memory Companion and LipSync video generation.

Demo Video

Watch the demo video

Prerequisites

• Node.js (v20 or higher)
• npm
• Inworld AI API Key
• Voice Clone & Get VoiceId
• Runway ML Account email (for LipSync Generation)
• UseAPI Token (for LipSync Generation)
• UseAPI & LipSync setup (for LipSync Generation)
• Runway ML API Key (for Memory Companion)

Get Started

Step 1: Clone the Repository

git clone https://github.com/inworld-ai/living-memories-node.git
cd living-memories-node

Step 2: Install Dependencies

npm install

Note: This project uses @inworld/runtime v0.8.0. Make sure to run npm install to get the correct version.

Step 3: Configure Environment Variables

Duplicate env.example to .env in the root directory.

Get your API key from the Inworld Portal.

Step 4: Run the Application

npm run dev

Step 5: Access the Application

Open your browser and navigate to: http://localhost:3000

You'll be automatically redirected to the Memory Companion page.

Repo Structure

Living-Memories_Node/
├── src/              # Source code
│   ├── server.ts     # Express server with Inworld Runtime integration
│   └── tsconfig.json # TypeScript configuration
├── public/           # Static assets (HTML, CSS, JavaScript)
│   ├── lipsync.html  # LipSync generation interface
│   ├── lipsync.js    # LipSync client logic
│   ├── memory.html   # Memory Companion interface
│   ├── memory.js     # Memory Companion client logic
│   └── style.css     # Shared styles
├── uploads/          # User-uploaded images
├── README.md         # Documentation
├── package.json      # Dependencies
└── LICENSE           # MIT License

Features

1. Memory Companion (/memory)

1. Enter your Inworld API Key, Voice ID, and Runway API Key
2. Enter your Companion Configuration
3. Upload an image you want to use to generate the video
4. (Optional) Add a text prompt to guide the animation
5. Generate an animated video from the image
6. Press and hold the button in the Step 3 area to record your voice
7. Chat with your companion using natural voice conversations

2. LipSync Generation (/lipsync)

1. Enter your Inworld API Key, Voice ID, UseAPI Token, and Runway Email that is linked with your UseAPI account
2. Upload a portrait image
3. Enter text for the character to speak
4. Generate TTS audio using Inworld's voice synthesis. Play it before generating the video to make sure that is the audio you want.
5. Generate a lip-synced video

Troubleshooting

Bug Reports: GitHub Issues

General Questions: For general inquiries and support, please email us at support@inworld.ai

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines on how to contribute to this project.

License

This project is licensed under the MIT License - see the LICENSE file for details.