Speaker speech segmentation method and device, electronic equipment and storage medium
A speaker and voice technology, which is applied in the field of speaker voice segmentation, can solve the problems of low segmentation accuracy and large amount of calculation, and achieve the effects of alleviating calculation pressure, reducing data volume, and improving segmentation accuracy
Pending Publication Date: 2020-11-03
CHINA MOBILEHANGZHOUINFORMATION TECH CO LTD +1
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The purpose of the embodiment of the present invention is to provide a speaker voice segmentation method, device, electronic equipment and storage medium, which can extract the optimal Mel-frequency cepstrum feature, and solve the problems of large amount of calculation and low segmentation accuracy
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0051] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, various implementation modes of the present invention will be described in detail below in conjunction with the accompanying drawings. However, those of ordinary skill in the art can understand that, in each implementation manner of the present invention, many technical details are provided for readers to better understand the present application. However, even without these technical details and various changes and modifications based on the following implementation modes, the technical solution claimed in this application can also be realized.
[0052] The embodiment of the present invention relates to a speaker's voice segmentation method. The core of this embodiment is to extract the Mel-frequency cepstrum feature of the speaker's voice, and select multiple Mel-frequency cepstrum features that are not repeated for clustering operation And segment ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The embodiment of the invention relates to the technical field of speech processing, and discloses a speaker speech segmentation method. The method comprises the following steps: carrying out silenceremoval when initial speaker speech comprises silence to obtain standard speaker speech, extracting Mel frequency cepstrum characteristics of the standard speaker speech to obtain a Mel frequency cepstrum characteristic set, selecting Mel frequency cepstrum characteristics of different dimensions from the Mel frequency cepstrum characteristic set, obtaining a Mel frequency cepstrum characteristicsubset, and performing clustering segmentation on the Mel frequency cepstrum characteristic subset to obtain speeches of different speakers. The invention further provides a speaker speech segmentation device, electronic equipment and a computer readable storage medium. According to the invention, the problems of large calculation amount and low segmentation accuracy in the speech segmentation process can be effectively solved.
Description
technical field [0001] Embodiments of the present invention relate to the technical field of speech processing, and in particular to a method, device, electronic equipment and readable storage medium for speaker speech segmentation. Background technique [0002] With the continuous improvement of informatization and automation in various industries in society, people's demand for accurate data is getting higher and higher. Taking the recording of door-to-door service as an example, the company that provides the service wants to check whether employees use standardized terms in the door-to-door service. During detection, it is necessary to accurately segment the voices of two people in the recording to ensure that there is only one speaker in a segment of voice. Segmentation is important. [0003] However, the inventors found that there are the following disadvantages when segmenting the speaker's voice: the calculation of the Mel frequency cepstrum features of each frame in...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G10L21/0272G10L21/0308G10L25/24
CPCG10L21/0272G10L21/0308G10L25/24
Inventor 单彦会郭晗暄张俊杰孟越涛李娜罗红
Owner CHINA MOBILEHANGZHOUINFORMATION TECH CO LTD