Koichi Shinoda

Papers

Total Citations

H-Index

About

Koichi Shinoda is a leading researcher in robust speech recognition and human-robot interaction, with a focus on developing systems that perform reliably in real-world, noisy environments. His major contributions center on addressing the challenge of nonstationary sudden noise—common in home settings—through the innovative use of factorial hidden Markov models (FHMMs). In his foundational 2007 work, cited 17 times, Shinoda demonstrated how FHMM architectures, trained on clean speech, can effectively model and compensate for unpredictable acoustic disturbances, significantly improving recognition accuracy in domestic spaces. This work has been instrumental in advancing speech interfaces for smart homes and assistive technologies. Beyond audio processing, Shinoda has explored affective computing, notably in a 2016 study on boredom recognition from spontaneous behaviors during multiparty human-robot interactions, earning 5 citations for its novel approach to social robotics. His research bridges robust signal processing and human-centered AI, with cumulative citations reflecting its impact on both theoretical frameworks and practical applications. Shinoda’s contributions are essential for students and engineers seeking to build resilient, context-aware voice systems and empathetic robots.