About Reseach

Music informatics

Structure analysis of music

We are working on hierarchical structure analysis of music, similarity detection, theme-variation relationship analysis, and harmony analysis using symbolic music data.
In addition to rule-based methods that take into account musical knowledge, we are also exploring methods that use cognitive music theory such as GTTM and IRM, and machine learning techniques (e.g. Word2Vec, LSTM generative model.)

Sound analysis of singing voice, musical instruments

In order to clarify the properties of songs and instrumental sounds, we are analyzing the acoustic characteristics using the RWC music database and the UTAU library.

Music cognition

We are investigating how humans listen to and play music. We are conducting psychological experiments and analysis of audio and music data.

Music information retrieval

We use computers to quantify information about music, and use it to recognize or generate musical elements.

Multimodal interaction with sounds and music

Sound Perception

We are conducting research to elucidate how humans hear and understand the various sounds that exist around us.

Sound and music perception by deaf and hard-of-hearing people

We are studying what characteristics and features of sound are used by hearing impaired people with residual hearing to understand music and environmental sounds, and what patterns and individual differences exist in this understanding.

Design of sound

We are exploring how to create sounds as a means of communicating information to people, such as making data audible so that it can be expressed and understood as sound, and designing alarm sounds that people can easily hear.