Virtual Human for Chatting

Live2D Virtual Human for Chatting based on Unity

Project introduction

Video demonstration: https://www.bilibili.com/video/BV1CW4y1X7PK

Project deployment

API application

• This project does not provide public API Key, please apply for it yourself.

  1. Azure: API documentation and set Region to eastasia
  2. OpenAI: API documentation
  3. APISpace: API documentation

• After applying, please fill in the API Key in the setting UI, and the project will store this data persistently.

environment

• Unity 2021.3.0

resolution

• It is recommended to set the resolution under Editor to 1920*1080
• The default setting is 1920*1080 under Runtime

Unity package import

• OpenCVPlusUnity: Processes image-related operations, used for face detection and other modules
• Newtonsoft.Json: JSON related operations such as serialization and deserialization
  • Add Package from git URL : com.unity.nuget.newtonsoft-json

Technical documentation

If you are interested in the implementation of this project, we have recorded some development and learning processes. FYI: Technical Documentation

license

• The code for this project is licensed under the MIT License.

• Models and art assets are owned by their respective authors and must be used with the permission of their respective authors.

Quote

If you find this project useful in your work, please consider citing:

@misc{VHC,
  author = {Tianyu and Si, Haoyu and Gao, Lingjie and Wang},
  title = {Live2D Virtual Human for Chatting based on Unity},
  year = {2023},
  publisher = {GitHub},
  journal = {https://github.com/Navi-Studio/Virtual-Human-for-Chatting},
  school = {Nanjing University}
}

Acknowledgments

We are very grateful to the artist Strawberry Milk Rabbit for providing the "Peach Blossom Cheese Motoko" Live2D model, and Yuzuyaowo for providing the exquisite background original painting. Many thanks to the above authors for their excellent work.

Thanks to [email protected] for providing the OpenAI API proxy, which allows this project to directly call OpenAI services in the domestic network environment.