Masuda-lab.

we investigated how the attractiveness of individual faces influences the perceived attractiveness of a group.

we improved the efficiency of model development and enhanced malware detection accuracy by performing parallel hyperparameter tuning for machine learning models.

In this study, we proposed a method to estimate restaurant congestion and calculate waiting times using live camera footage from the Internet and object detection technology.

In this study, we investigated the quality of programs automatically generated by large language model (LLM) by applying code metrics—such as the number of code steps and complexity—and static analysis tools.

In this study, we developed a simplified work schedule management system using “LibreOffice Calc, ” (※1) a free open-source software (OSS).

In this study, we examined the communication characteristics of network-based video transmission using Network Device Interface (NDI).

This study investigated methods for efficiently generating voice clones from a small amount of speech data.

This study aimed to predict future changes in medical expenditures and examine regional disparities using machine lerarning models.

Our laboratory participated in the 2023 Yokohama Festival of Tokyo City University and operated a shaved ice booth.

A demonstration of voice cloning technology was conducted during the Open Campus event held in August 2022. The system enabled participants to experience the generation of synthetic speech that closely approximated their own vocal characteristics.