Yuexian Zou

Papers

Total Citations

H-Index

About

Yuexian Zou is a leading researcher in acoustic signal processing and deep learning for audio analysis, with a focus on robust speech enhancement, direction-of-arrival (DOA) estimation, and intelligent sound event detection. Her work bridges the gap between traditional signal processing and modern neural network approaches, particularly for challenging real-world environments such as low signal-to-noise ratio (SNR) or high reverberation conditions. In a highly cited 2018 study, Zou introduced AICDS, an infant crying detection system using a lightweight convolutional neural network, demonstrating practical applications of deep learning in healthcare and smart home contexts. She has also made significant contributions to DOA estimation by leveraging acoustic vector sensor (AVS) cues to improve accuracy and robustness in small microphone arrays—a critical need for service robotics. Another notable achievement is her development of a nonlinear soft masking approach for target speech enhancement using a single AVS, where she derived inter-sensor data ratio (ISDR) models in the time-frequency domain. With over 20 citations across her most prominent works, Zou’s research is highly regarded for its innovation in combining theoretical signal models with data-driven methods, offering scalable solutions for real-world audio processing challenges.