Recording terminal clustering method based on voice time-frequency transform features and integral linear programming

A technique of integer linear programming and time-frequency transformation, which is applied in speech analysis, instruments, etc., and can solve the problem of reducing the universality of the method.

Inactive Publication Date: 2018-11-30
SOUTH CHINA UNIV OF TECH
View PDF10 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If the above-mentioned prior information cannot be obtained, the unive

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
  • Recording terminal clustering method based on voice time-frequency transform features and integral linear programming
  • Recording terminal clustering method based on voice time-frequency transform features and integral linear programming
  • Recording terminal clustering method based on voice time-frequency transform features and integral linear programming

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0152] figure 1 It is a flowchart of a recording terminal clustering method using voice time-frequency transformation characteristics and integer linear programming disclosed in the present invention, wherein the recording terminal includes smart phones, MP3, MP4, recording pens, notebook computers, tablet computers, Wear a recording device or other type of terminal recording device. In this embodiment, taking a smart phone as an example, a method for clustering recording terminals using voice time-frequency transformation features and integer linear programming is specifically introduced, including the following steps:

[0153] Step 101, read in recording samples, the recording samples are recorded by smart phones of various brands and models, and the format is WAV.

[0154] Step 102: Perform preprocessing on the recording samples read in step 101. The preprocessing operation includes steps such as pre-emphasis, framing, windowing, and discrete Fourier transform of speech. ...

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 recording terminal clustering method based on voice time-frequency transform features and integral linear programming. The method comprises the following steps: a, carrying out preprocessing on read recording; b, extracting GTCC features from each frame of voices; c, constructing a depth self-coding network with a bottleneck layer, and extracting bottleneck features; d, constructing gaussian supervector features; e, carrying out spectrogram feature extraction on the voice obtained after preprocessing in step a; f, splicing the spectrogram features in step e to the gaussian supervector in step d, thus obtaining voice time-frequency transform features describing the features of the recording terminal finally; and g, carrying out clustering on the voice time-frequency transform features of all the recording samples by adopting the integral linear programming algorithm, wherein the clustering result is taken as the final classification basis. The feature extraction and clustering of the method provided by the invention are unsupervised, and compared with the existing supervising method, the method provided by the invention has better universality.

Description

technical field [0001] The present invention relates to the technical fields of digital voice processing and recording terminal source forensics, in particular to a recording terminal clustering method based on speech time-frequency transform (Speech Time-Frequency Transform, STFT) features and integer linear programming (Integer Linear Programming, ILP). Background technique [0002] With the popularity of portable recording devices, especially smartphones, the amount of audio data recorded by people has exploded. How to effectively identify the recording terminal of the above-mentioned audio data is one of the research hotspots of digital audio forensics technology at present. [0003] At present, the research work on source forensics of recording terminals mainly starts from the perspective of identification, that is, the identification or confirmation of recording terminals. However, in this process, it is necessary to predict the brand, model and other prior informatio...

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): G10L25/45G10L25/18G10L25/03G10L25/24G10L25/27G10L25/48
CPCG10L25/03G10L25/18G10L25/24G10L25/27G10L25/45G10L25/48
Inventor 李艳雄张雪张聿晗李先苦
Owner SOUTH CHINA UNIV OF 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