Belhedi Wiem

Papers

Total Citations

H-Index

About

Belhedi Wiem is a researcher focused on advancing speech processing and computational intelligence, with key contributions in single channel speech separation (SCSS) and automatic object recognition. Her work addresses the critical challenge of isolating clean speech from mixed audio signals in noisy environments—a task essential for applications like smart home voice commands and human-robot interaction. Her most cited paper, "Unsupervised single channel speech separation based on optimized subspace separation" (2017, 12 citations), introduces an innovative approach that enhances separation quality without requiring labeled training data. Building on this, her 2020 study on phase-aware subspace decomposition (5 citations) further refines SCSS by incorporating phase information, improving performance in real-world acoustic conditions. Additionally, her 2021 exploration of computational intelligence for automatic object recognition in vision systems (2 citations) demonstrates versatility across domains. Wiem’s work is notable for its practical impact on post-processing in communication systems, offering robust solutions for challenging acoustic environments. Her research continues to influence the development of more reliable and intelligent audio and vision technologies.