Kernel-function-based dimension reduction method of audio feature signal

An audio signal and audio feature technology, applied in the field of audio feature signal processing, can solve problems such as errors and high complexity of dimensionality reduction algorithms, and achieve the effects of easy programming, simple theory, and improved processing speed

Active Publication Date: 2018-12-21
KUNMING UNIV OF SCI & TECH
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Most of these dimensionality reduction algorithms are highly complex, to discard part of the characteristic signals to achieve the purpose of dimensi

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
  • Kernel-function-based dimension reduction method of audio feature signal
  • Kernel-function-based dimension reduction method of audio feature signal
  • Kernel-function-based dimension reduction method of audio feature signal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0038] Such as Figure 1-3 As shown, a method for dimensionality reduction of audio feature signals based on kernel functions, the specific steps are:

[0039] (1) Audio signal collection: collect audio signals and obtain audio samples.

[0040] (2) Audio signal preprocessing: convert the analog signal in the collected audio samples into a digital signal, and write the digital signal into a WAV file. Filter, pre-emphasize, and frame the digital signal to be written into the WAV file.

[0041] (3) Feature parameter extraction: extract high-dimensional feature parameters from the linear predictive coefficient (LPC), linear predictive cepstral coefficient (LPCC), and Mel frequency cepstral coefficient (MFCC) in the processed digital signal.

[0042] (4) Construction of dimensionality reduction model: send the above-mentioned extracted feature parameters...

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, which belongs to the technical field of audio signal processing, relates to a kernel-function-based dimension reduction method of an audio feature signal. Dimension reduction processingis carried out on a feature parameter of an audio signal; the required dimension reduction effect is realized while the audio feature information amount is not abandoned; and then last dimension reduction data are displayed visually and comparative analysis is carried out on the last dimension reduction data and results obtained by other audio feature parameter dimension reduction methods. According to the invention, for dimension reduction of the audio feature parameter, dimension reduction processing is mainly carried out on a linear prediction coefficient, a linear prediction cepstral coefficient and a Mel frequency cepstral coefficient in an audio coefficient domain and the data results after dimension reduction are displayed visually. The audio feature dimension reduction processingcan be applied to broadcast signal monitoring and quick identification processing of audio signals. The algorithm is simple and the nonlinear kernel function is used for expressing a mapping relationship between Gaussian observation space and hidden space, thereby overcoming defects of usage range limitation and poor dimension reduction effect of the linear mapping method.

Description

technical field [0001] The invention relates to a dimensionality reduction method of an audio feature signal based on a kernel function, and belongs to the technical field of audio feature signal processing. Background technique [0002] In order to realize the management and control of wireless audio broadcasting, and carry out safe and efficient real-time monitoring and screening of audio broadcasting, the rapid processing of audio information is related to the process speed of the entire process, and the dimensionality reduction processing of audio characteristic signals is the core of audio information processing. Efficiency and credibility must also become problems to be solved urgently. At present, most of the dimension reduction methods for audio feature signals mainly include local preservation projection method, multi-dimensional scaling method, local linear embedding method, principal component analysis method, etc. Most of these dimensionality reduction algorithm...

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): G10L25/24G10L25/03G10L25/45G10L19/00
CPCG10L19/00G10L25/03G10L25/24G10L25/45
Inventor 龙华杨明亮邵玉斌杜庆治
Owner KUNMING UNIV OF SCI & TECH
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap