Elsevier

Speech Communication

Volume 127, March 2021, Pages 82-91
Speech Communication

Speech signal processing on graphs: The graph frequency analysis and an improved graph Wiener filtering method

https://doi.org/10.1016/j.specom.2020.12.010Get rights and content
Under a Creative Commons license
open access

Highlights

  • Proposed a graph k-shift operator to build up a new directed graph topology for speech signals.

  • Construct a new graph Fourier basis based on the singular eigenvectors of the defined graph k-shift operator to perform GFT.

  • Proposed an improved graph Wiener filtering method based on the defined graph k-shift operator and investigated the gain function of the improved graph Wiener filtering method in graph frequency domain.

  • Performance of the improved graph Wiener filtering method performs better especially under low SNR conditions.

Abstract

In the paper, we investigate a graph representation of speech signals and graph speech enhancement technology. Specifically, we first propose a new graph k-shift operator Ck to map speech signals into the graph domain and construct a novel graph Fourier basis by using its singular eigenvectors for speech graph signals (SGSs). On this basis, we propose an improved graph Wiener filtering method based on the minimum mean square error (MMSE) criterion to suppress the noise interference in noisy speech. Comparing with the traditional methods in DSP and the existed graph Wiener filtering methods by applying graph shift operators in GSP, our numerical simulation results show that the performance of the proposed method outperforms that of these methods in terms of both average SSNR and mean PESQ score. Moreover, the computational complexity of the proposed method is much lower than that of the existed graph Wiener filtering methods and a little higher than that of classical methods in DSP.

Keywords

Graph signal processing
Graph Fourier transform
Graph Wiener filtering
Graph topology
Speech enhancement

Cited by (0)

Tingting Wang

Haiyan Guo

Xue Yan

Zhen Yang

This work has been supported by National Natural Science Foundations of China (No. 62071242, No. 61271335, No. 61901229, No. 61671252)