Speech Enhancement Based on Multi-Task Adaptive Knowledge Distillation

• Published in: Taiyuan li gong da xue xue bao = Journal of Taiyuan University of Technology Vol. 55; no. 4; pp. 720 - 726
• Main Authors: ZHANG Gangmin, LI Yarong, JIA Hairong, WANG Xianxia, DUAN Shufei
• Format: Journal Article
• Language: English
• Published: Editorial Office of Journal of Taiyuan University of Technology 01-07-2024
• Subjects: knowledge distillation; multi-task adaptive learning; speech enhancement; weighted loss function
• DOI: 10.16355/j.tyut.1007-9432.20230259
• ISSN: 1007-9432

Purposes In order to solve the computational cost problem of complex model in time and hardware, and improve the performance of speech enhancement algorithm, a speech enhancement algorithm using multi-task adaptive knowledge distillation is proposed. Methods First, the idea of knowledge distillation is adopted to solve the problems that the existing speech enhancement model is too large, has many parameters, and has high calculation cost. Second, the differences between different time-frequency units are fully considered, and the weighting factor is introduced to optimize the traditional loss function to improve the network performance of students. In order to avoid the uncertainty of teacher network prediction affecting the performance of student network, the knowledge distillation network of multi-task adaptive learning is built to better utilize the correlation between different tasks to optimize the model. Findings The simulation results show that the proposed algorithm can effectively improve the performance of speech enhancement model while reducing the number of parameters and shortening the calculation time.