Max Robotics — Global Service Robot Directory + US Certification

About

Masashi Unoki is a leading researcher in speech emotion recognition and auditory-inspired signal processing. His work focuses on developing robust computational models that mimic the human auditory system to analyze emotional cues in speech, enabling more natural human-robot interaction. Unoki’s major contributions include pioneering the use of multi-resolution modulation-filtered cochleagram features and advanced deep learning architectures—such as 3D convolutions and attention-based sliding recurrent networks—to capture the temporal dynamics of emotion from speech. His most cited paper (2020, 83 citations) demonstrates how auditory front-ends can effectively track emotional intensity and fundamental frequency, significantly improving emotion recognition accuracy. He has also advanced dimensional emotion recognition (DER) by integrating modulation spectral features with recurrent neural networks, allowing robots to continuously track emotional states over time. With over 130 total citations, Unoki’s work bridges auditory perception and machine learning, offering practical solutions for affective computing. His research is particularly notable for its emphasis on biologically plausible feature extraction, which enhances system robustness in noisy environments. Unoki’s achievements position him as a key figure in developing emotionally intelligent robots that can understand human intentions through natural speech.