We are surrounded by audio sources such as speech of a man, music from a radio, or knocks on a door.
They contain full of diversities and are interfering with each other.
I am interested in the technology that enables computers to analyze these audio signals,
or observations in a natural environment,
based on signal processing and machine learning techniques.
- Application of Bayesian probabilistic models to microphone array processing Demo page Japanese audio
- Music audio signal processing, Co-player music robot capable of a musical ensemble with human musicians
- Measurement and analysis of frogs' inter-callings
- Analysis of human (singing) voice