The hoyoGPT model is an innovative musical composition model developed based on the Tunesformer architecture. Trained on the MiHoYo Music Remix Piano Dataset, this model is specifically tailored for generating music inspired by user-created content within the MiHoYo gaming community. Currently, the training data is limited to sliced ABC musical notations derived from remixed tunes in MiHoYo's games, specifically focusing on "Genshin Impact" and "Honkai: Star Rail" Leveraging the tunesformer model, the MiHoYo Music Generation Model aims to capture the essence of the unique musical styles present in the MiHoYo gaming universe. This model represents a fusion of cutting-edge technology and artistic creativity, providing a platform for the generation of novel and captivating musical compositions within the context of MiHoYo's gaming experiences.

Maintenance

GIT_LFS_SKIP_SMUDGE=1 git clone git@hf.co:MuGeminorum/hoyoGPT

Mirror

https://www.modelscope.cn/models/MuGeminorum/hoyoGPT

Reference

[1] https://github.com/MuGeminorum/hoyoGPT
[2] https://huggingface.co/sander-wood/tunesformer

Downloads last month
0
Unable to determine this model’s pipeline type. Check the docs .

Dataset used to train MuGeminorum/hoyoGPT

Space using MuGeminorum/hoyoGPT 1

Collection including MuGeminorum/hoyoGPT