Intelligent Speech Processing Technology

Liejun Wang is a Professor in the School of Computer Science and Technology at Xinjiang University. His research interests include multimedia information processing, computer vision, and intelligent information systems. He has led more than 50 research projects, including national-level and provincial-level projects, and has published more than 170 academic papers. He has received several awards and honors, including the Medal for Development and Construction of Xinjiang, the Baosteel Excellent Teacher Award, the Tianshan Talents High-Level Leading Talent Program, and the Second Prize of the Natural Science Award of Xinjiang Uygur Autonomous Region.

Yinfeng Yu is an Associate Professor at Xinjiang University. He received his PhD degree in Computer Science and Technology from Tsinghua University in 2023. His research interests include artificial intelligence, intelligent robotics, speech interaction, remote sensing image analysis, and multimedia information processing. He has published papers in leading international conferences such as IJCAI and ICLR and has served as a reviewer for major conferences including ICASSP, CVPR, ICCV, and ECCV.

Xiaoming Tao is a Tenured Professor in the Department of Electronic Engineering at Tsinghua University and a recipient of the National Science Fund for Distinguished Young Scholars. Her research interests include wireless multimedia communication, multimedia computational communication, and semantic communication. She has received several major awards, including the China Young Women Scientist Award, the China Youth Science and Technology Award, the Xplorer Prize, and national-level science and technology awards.

Introduction.- The Development History of Speech Technology.- Overview of Speech Technology.- Speech Emotion Recognition.- MFHCA: Enhancing Speech Emotion Recognition Via Multi-Spatial Fusion and Hierarchical Cooperative.- Emotion Recognition in Speech via Progressive Channel Querying.- BSS-CFFMA¿BSS-CFFMA: Cross-Domain Feature Fusion and Multi-Attention Speech Enhancement Network based on Self-Supervised Embedding.- BSP-MPNet: Magnitude-Phase Dual-Path Speech Enhancement Network based on Self-Supervised Embedding and Perceptual Contrast Stretch Boosting.- Speech Recognition.- Speech Synthesis.- Conclusion.