Yoshikazu Miyanaga

Papers

Total Citations

H-Index

About

Yoshikazu Miyanaga is a leading researcher in robust speech processing, embedded systems, and assistive robotics, with a focus on real-time signal enhancement for human-machine interaction. His most cited work, “Noisy speech training in MFCC-based speech recognition with noise suppression toward robot assisted autism therapy” (2017, 21 citations), pioneers the integration of noise-robust automatic speech recognition into socially assistive robots for children with Autism Spectrum Disorder (ASD). By combining Mel-frequency cepstral coefficient (MFCC) training with adaptive noise suppression, Miyanaga’s system enables robots to understand speech in challenging acoustic environments, directly supporting therapists in social communication interventions. He further advances real-time visual processing through “An Architecture for Real-Time Retinex-Based Image Enhancement and Haze Removal and Its FPGA Implementation” (2019, 12 citations), designing efficient hardware architectures that simultaneously enhance images and remove haze—critical for autonomous navigation and surveillance. His earlier work, “Robust speech communication and its embedded smart robot system” (2013), establishes foundational noise-robust speech detection and rejection algorithms implemented on low-power LSIs. Across his career, Miyanaga’s contributions bridge algorithmic innovation with practical embedded deployment, achieving measurable impact in assistive technology and real-time signal processing.