I got my master degree at University of Tsukuba and the Lab. for Sound and People with Computing.

Before the master program, I got my bachelor’s degree at Dalian University of Technology, majored in Computer Science and Technology(Japanese Intensive). At that time, I becomed interested in music and started to imagine creating music by Artificial Intelligence.

My research topics are about Lyrics/Music Generation&Understanding, Multi-modality Data Alignment, Multilingual Research. Including technologies such as LLM, computer music, AI/ML, signal processing. My dream is to explore the intrinsic consistency of music, human languages, and mathematics. I’d like to fuse the inspiration by human with the “learning” by computer to creative fantasic AI music.

Research Projects

Lyrics Generation
Title: 《Modeling the Structural and Semantic Features for Japanese Lyrics Generation of J-pop Songs》

Abstract: we propose a Japanese lyrics generation model named Okashi that can generate lyrics with clear structure, high text quality, and conformity to human lyrics writing principles.

Symbolic Music Generation Title: 《Chord Conditional Automatic Songwriting System with Phrase-Motif Level Alignment》

Abstract: This research aims to develop an automatic song generation system that takes chord progression as input to generate lyrics and melody that are aligned in terms of structure and musicality.

🎼To sum up, computational musicology is an interesting but challenging interdisciplinary research field but have not been explored well. We need more effects! My future research programs show as below:

• Lyrics Translation
• Musical features analysis/Alignment
• Any-modality-to-Music Generation
• To be continued…