DialogLab

Paper | 30s Preview | Video Figure | Web Demo

DialogLab is a tool for configuring and running multi-agent conversations, with a React + Vite client and an Express server.

Prerequisites

• Node.js 18+ (client build and deploy scripts recommend Node 23)
• npm 8+

Repository layout

• client/: React UI (Vite). Dev server on port 5173. Proxies API to the server.
• server/: Express API/Web server. Listens on port 3010.

Quick start (development)

1. Install dependencies

• cd client && npm install
• cd server && npm install

2. Configure environment (server/.env) Create a .env file in server/ with the keys you intend to use. Example:

# Core
NODE_ENV=development

# LLM providers (choose one or both)
GEMINI_API_KEY=your-gemini-api-key
API_KEY_LLM=your-openai-api-key

# Defaults (optional)
DEFAULT_LLM_PROVIDER=gemini   # or openai
DEFAULT_OPENAI_MODEL=gpt-4
DEFAULT_GEMINI_MODEL=gemini-2.0-flash

# Text-to-Speech (optional)
TTS_API_KEY=your-google-tts-api-key
TTS_ENDPOINT=https://eu-texttospeech.googleapis.com/v1beta1/text:synthesize

3. Run the server

cd server
node server.js

The server will start at http://localhost:3010.

4. Run the client

cd client
npm run dev

Open the client at http://localhost:5173.

Third-Party Components

TalkingHead (MIT License)

Portions of this project’s code are adapted from the open-source project TalkingHead,
© 2024 Mika Suominen (met4citizen), and are used under the MIT License.
This applies to:

• client/public/libs/talkinghead.mjs

---

Three.js (MIT License)

This project uses Three.js and its example modules,
© 2010–present Three.js authors, licensed under the MIT License.
See https://github.com/mrdoob/three.js/blob/dev/LICENSE for details.

---

Ready Player Me Avatars (Custom License)

Example avatar files (client/public/assets/*.glb) were created using Ready Player Me
and are subject to Ready Player Me’s Terms of Use.
These assets are provided for demonstration purposes only and are not covered by this project’s open-source license.

Citation

If you use DialogLab in your research, please cite:

@inproceedings{dialoglab2025,
author = {Hu, Erzhen and Chen, Yanhe and Li, Mingyi and Phadnis, Vrushank and Xu, Pingmei and Qian, Xun and Olwal, Alex and Kim, David and Heo, Seongkook and Du, Ruofei},
  title     = {DialogLab: Configuring and Orchestrating Multi-Agent Conversations},
  booktitle = {Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology (UIST '25)},
  year      = {2025},
  publisher = {Association for Computing Machinery},
  address   = {New York, NY, USA},
  doi       = {10.1145/3746059.3747696}
}