Voice activity detection in the presence of transient based on graph

EURASIP Journal on Audio, Speech, and Music Processing

卷:2023

Empirical Research

Chun-Xian Gao¹  Hui Liu¹  Xiao-Yuan Guo¹ 

[1] Department of Information and Communication Engineering, Xiamen University, Xiamen, China;

关键词: Voice activity detection;    Transients;    Time series complex networks;    Nonlinear dynamic characteristics;   

DOI  :  10.1186/s13636-023-00282-x

 received in 2022-06-23, accepted in 2023-03-25,  发布年份 2023

来源: Springer

【 摘 要 】

Voice activity detection remains a significant challenge in the presence of transients since transients are more dominant than speech, though it has achieved satisfactory performance in quasi-stationary noisy environments. This paper studies the differences between speech and transients in nonlinear dynamic characteristics and proposes a new method for accurately detecting speech and transients. Limited by algorithm complexity, previous research has proposed few detectors to model speech and transients based on contextual information and thus failing to detect transient frames accurately. To address this challenge, our study proposes to map features of audio signals to a time series complex network, a kind of graph data, analyzed by the Laplacian and adjacency matrix of graphs, then classified by the support vector machine (SVM) classifier. The proposed algorithm can analyze a more extended speech period, allowing the full utilization of contextual information of preceding and following frames. The experimental results show that the performance of this method has obvious superiority over other existing algorithms.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

40507_2023_167_Article_IEq56.gif

12942_2022_316_Article_IEq12.gif

10194_2023_1579_Article_IEq31.gif

10194_2023_1579_Article_IEq29.gif

Fig. 1

10194_2023_1579_Article_IEq26.gif

12942_2022_316_Article_IEq4.gif

Fig. 5

10194_2023_1579_Article_IEq24.gif

10194_2023_1579_Article_IEq22.gif

10194_2023_1579_Article_IEq18.gif

Fig. 5

Fig. 8

10194_2023_1579_Article_IEq13.gif

Fig. 15

Fig. 7

Fig. 11

Fig. 2

Fig. 4

Fig. 10

Fig. 14

Fig. 1

Fig. 9

10194_2023_1579_Article_IEq1.gif

Fig. 4

Fig. 8

Fig. 2

Fig. 3

Fig. 1

Fig. 13

41535_2022_516_Article_IEq34.gif

Fig. 1

41535_2022_516_Article_IEq31.gif

Fig. 1

Fig. 3

Fig. 10

Fig. 3

Fig. 4

Fig. 4

Fig. 9

41535_2022_516_Article_IEq16.gif

Fig. 2

Fig. 4

Fig. 1

Fig. 1

41535_2022_516_Article_IEq11.gif

Fig. 2

Fig. 6

Fig. 1

Fig. 2

Fig. 3

Fig. 3

13065_2022_881_Figi_HTML.gif

13065_2022_881_Figf_HTML.gif

13065_2022_881_Figc_HTML.gif

Fig. 3

Fig. 6