Nobuhisa Hirata
Papers
1
Total Citations
3
H-Index
1
About
Nobuhisa Hirata is a researcher whose work sits at the intersection of robotics, audio signal processing, and human-robot interaction. His primary research areas include multichannel noise reduction, speech extraction, and the integration of visual information to improve auditory perception in robotic systems. Hirata's most notable contribution is his pioneering approach to semi-blind speech extraction, where he leverages visual cues alongside noise statistics to enhance speech recognition accuracy in spoken-dialogue robots. His 2011 paper on this topic introduced a novel permutation solving method using a probability statistics model, specifically designed to handle realistic sound mixtures that combine point-source speech with diffuse noise. This work directly addresses one of the most challenging problems in robot audition: isolating a target speaker in noisy, dynamic environments. While his citation count is modest, the conceptual and methodological advances in his research have laid important groundwork for more robust, context-aware auditory systems in robotics, demonstrating a thoughtful integration of multimodal sensing for practical, real-world applications.
Research Focus
Key Achievements
Top Papers
- 1