Unlock instant, AI-driven research and patent intelligence for your innovation.

Method for extracting audio object from audio content based on projection

An audio and object technology, applied in the field of audio content processing, can solve problems such as incorrect position estimation

Inactive Publication Date: 2016-08-24
DOLBY LAB LICENSING CORP
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If the source separation technique is not used, the two objects will be considered as one object, which will make their position estimation incorrect

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 for extracting audio object from audio content based on projection
  • Method for extracting audio object from audio content based on projection
  • Method for extracting audio object from audio content based on projection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0051] According to an example embodiment of the present invention, one potential method for determining the diagonal values ​​of H is to set them according to the correlation matrix R. As mentioned above, the diagonal elements of R reflect the similarity between a pair of channels mapped to the projected space constructed by the column vectors of W (eg, Wx or Wy). Therefore, a higher similarity score indicates a higher likelihood that the same objects exist and can be recovered from these spaces. Therefore, it is reasonable to extract "more" objects from those spaces with higher similarity scores, that is, H can be represented by an appropriate function of R, namely:

[0052] H=f(R) (9)

[0053] Among them, the function f can be any function whose value does not decrease with the increase of the input value. For example, H could be a normalized R where the sum of the diagonal elements equals 1.

[0054] As mentioned above, the first channel and the second channel may be an...

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 relates to audio object extraction and discloses a method for extracting an audio object from audio content. The method includes the following steps that: a first projection space set is identified, wherein the first projection space set includes a first subset used for a first channel in a plurality of channels and a second subset used for a second channel in the plurality of channels; a first correlation set between the first channel and the second channel is determined, wherein each correlation in the first correlation set is corresponding to one projection space in the first projection space subset and one projection space in the second projection space subset; and the audio object is extracted from audio signals of the first channel based on the first correlations in the first correlation set and the projection spaces in the first subset which are corresponding to the first correlations, wherein the first correlations are larger than a first predefined threshold value. The invention also discloses a corresponding system and a computer program product.

Description

technical field [0001] Embodiments of the present invention generally contemplate audio content processing, and more particularly, relate to a method and system for extracting audio objects from audio content. Background technique [0002] Traditionally, audio content has been created and stored in a channel-based format. In channel-based formats, audio content is typically represented, stored, communicated, and distributed through the medium of channels. As used herein, the term "audio track" or "channel" refers to audio content that generally has a predefined physical location. For example, stereo, surround 5.1, surround 7.1, etc. are all channel-based formats for the audio content. Each channel corresponds to a fixed-position physical speaker. When multi-channel content is played back, multiple speakers create a real-time and immersive sound field that surrounds the listener. Recently, several traditional multi-channel systems have been extended to support new formats...

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
Patent Type & Authority Applications(China)
IPC IPC(8): H04S5/00H04S5/02
CPCH04S5/00H04S2400/03H04S2400/11H03H2021/0034G06F18/2134G06F17/15H03H21/00
Inventor 胡明清芦烈陈连武
Owner DOLBY LAB LICENSING CORP