Method and device for extracting target source from audios

A target source, audio technology, applied in speech analysis, instruments, etc., to achieve the effect of convenient processing and use

Active Publication Date: 2018-01-12
苏州音曼科技有限公司
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The first step in doing this is to extract the original vocals in the song separately from the accompaniment and vocals of the song. to extract out, become a problem

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for extracting target source from audios
  • Method and device for extracting target source from audios
  • Method and device for extracting target source from audios

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0020] see figure 1 As shown, a method for extracting target source from audio, including steps:

[0021] Perform time-frequency transformation on the collected audio signal frame by frame, transform the time-domain signal into a frequency-domain signal, and use a window function to segment the frequency-domain signal to form the first signal a and the second signal b;

[0022] Iteratively calculate the virtual angle θ of the virtual source corresponding to each frequency point of the first channel signal and the second channel signal of each frame frequency domain signal at a given frequency k ab (k);

[0023] Compare the virtual angle θ ab (k) and the size of the predetermined a...

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

No PUM Login to view more

Abstract

The invention discloses a method and device for extracting a target source from audios. The method comprises the steps of performing time-frequency transforming on collected audio signals frame by frame, converting time-domain signal into frequency-domain signals, and utilizing a window function for cutting the frequency-domain signals to form two ways of signals; traversing and calculating the virtual included angle of a virtual source corresponding to frequency points of the two ways of signals of each frame of the frequency-domain signal under the given frequency; comparing the virtual included angle with a preset angle threshold value, and according to a comparative result, adopting one signal as a target source signal and extracting the frequency-domain signals of the target source signal to be stored; utilizing time-frequency inverse transforming for converting the stored frequency-domain signals of the target source signal into time-domain signals, and outputting the target source time-domain signals. The target source signals are separated from the audio signals.

Description

technical field [0001] The invention relates to the technical field of audio signal processing, in particular to a method and device for extracting a target source from audio. Background technique [0002] At present, most of the singing scoring systems in the KTV market are scored based on the pitch fluctuation or volume of the singing, and cannot really be scored according to the voice of the singer. This low-precision scoring system is increasingly unable to meet the needs of consumers. need. Imagine that a person who sings very well and a person who sings not so well have the same score, or a person who sings badly because of the volume has a higher score. This kind of scoring greatly reduces the enthusiasm of some people. Therefore, it is very important to improve the KTV scoring system. In order to improve the KTV scoring system and make the scoring more accurate, we can compare the vocals of consumers singing in KTV with the original vocals in the song. The higher t...

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
no application Login to view more
IPC IPC(8): G10L21/0272G10L25/48G10L25/78
Inventor 郑羲光尚梦宸刘飞
Owner 苏州音曼科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products