Media Science Program

HOME > Major Subject > Media Science Program > Yasunari OBUCHI

Yasunari OBUCHI


Yasunari OBUCHI

Department/Research Field

Media Science program/Acoustics and Speech Processing

Research content
We study the process of analyzing the sound input, extracting various information from it, and modifying it for particular purposes. The research themes of our laboratory include separation of voice and noise, estimation of speaker’s emotion from voice, classification of environmental and mechanical sounds, and developing multimedia performance system based on sound analysis.
Yasunari OBUCHI
Example of noise suppression: (upper-left) waveform of noisy signal (lower-left) waveform of noise suppressed signal (upper-right) spectrogram of noisy signal (lower-right) spectrogram of noise suppressed signal

Research theme

1. Robust speech recognition
2. Speech enhancement
3. Emotion recognition

Research keywords
Speech , Acoustics , Multimedia
Lab link