An MMSE graph spectral magnitude estimator for speech signals residing on an undirected multiple graph

EURASIP Journal on Audio, Speech, and Music Processing

Methodology

Qiquan Zhang¹  Zirui Ge²  Haiyan Guo²  Tingting Wang²  Zhen Yang² 

[1] Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore;School of Communication and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing, China;

关键词: Graph signal processing;    K;    Graph representation;    MMSE;    Speech enhancement;   

DOI  :  10.1186/s13636-023-00272-z

 received in 2022-03-26, accepted in 2023-01-06,  发布年份 2023

来源: Springer

【 摘 要 】

The paper uses the K-graphs learning method to construct weighted, connected, undirected multiple graphs, aiming to reveal intrinsic relationships of speech samples in the inter-frame and intra-frame. To benefit from the learned multiple graphs’ property and enhance interpretability, we study the spectral property of speech samples in the joint vertex-frequency domain by using the new graph weight matrix. Moreover, we propose the representation of minimum mean-square error (MMSE) graph spectral magnitude estimator for speech signals residing on undirected multiple graphs. We use the MMSE graph spectral magnitude estimator to improve speech enhancement performance. The numerical simulation results show that the proposed method outperforms the existing methods in graph signal processing (GSP) and the baseline methods for speech enhancement in discrete signal processing (DSP) in terms of PESQ, LLR, output SNR, and STOI results. These results also demonstrate the validity of the learned multiple graphs.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

12302_2023_718_Article_IEq20.gif

Fig. 6

Fig. 3

Fig. 5

Fig. 8

Fig. 2

Fig. 2

Fig. 6

Fig. 2

Fig. 4

Fig. 1

Fig. 1

Fig. 1

Fig. 8

Fig. 5

Fig. 4

Fig. 2

Fig. 3

12302_2023_718_Article_IEq4.gif

Fig. 2

Fig. 4

Fig. 2

40708_2023_184_Article_IEq16.gif

Fig. 1

Fig. 5

Fig. 1

Fig. 6

Fig. 1

Fig. 2

Fig. 3

Fig. 7

Fig. 7

Fig. 1

Fig. 1

Fig. 10

Fig. 5

Fig. 5

Fig. 8

Fig. 7

Fig. 4

Fig. 6

Fig. 4

Fig. 3

Fig. 6

Fig. 2

Fig. 3

Fig. 1

40854_2022_419_Article_IEq15.gif

Fig. 6

Fig. 2

Fig. 2

Fig. 1

Fig. 5

Fig. 1

Fig. 5

Fig. 4

Fig. 1

Fig. 2

Fig. 3

Fig. 8

Fig. 4

Fig. 2

Fig.1

Fig. 1

Fig. 2

Fig. 3

13690_2023_1046_Article_IEq7.gif

Fig. 3

Fig. 5

Fig. 8

13690_2023_1046_Article_IEq1.gif

Fig. 2

Fig. 7

Fig. 2

Fig. 1

Fig. 6

Fig. 7

Fig. 1

12936_2023_4470_Article_IEq1.gif

Fig. 5

Fig. 6

Fig. 1

Fig. 5

Fig. 2

Fig. 6

Fig. 4

Fig. 9

Fig. 2

Fig. 5

Fig. 5

Fig. 2

Fig. 4

Fig. 1

Fig. 4

Fig. 4

Fig. 6

Fig. 3

Fig. 2

Fig. 5

Fig. 1

Fig. 2

Fig. 1

Fig. 5

Fig. 2

Fig. 4

Fig. 1

40854_2022_440_Article_IEq31.gif

Fig. 1

Fig. 2

Fig. 6

Fig. 4

Fig. 4

Fig. 2

Fig. 2

Fig. 3

Fig. 2

Fig. 2

Fig. 8

Fig. 1

Fig. 1

Fig. 4

Fig. 6

Fig. 3

Fig. 1

Fig. 9

Fig. 1

Fig. 2

Fig. 22

Fig. 4

Fig. 1

Fig. 7

40854_2022_440_Article_IEq23.gif

Fig. 1

Fig. 8

Fig. 3

Fig. 4

40854_2022_440_Article_IEq15.gif

Fig. 1

Fig. 6

Fig. 2

Fig. 1

Fig. 2

Fig. 5

Fig. 1

Fig. 2

Fig. 5

Fig. 1

Fig. 2

13690_2023_1037_Article_IEq1.gif

Fig. 4

40708_2023_185_Article_IEq64.gif

Fig. 2

Fig. 5

Fig. 5

40249_2023_1061_Article_IEq17.gif