Supercharge Your Innovation With Domain-Expert AI Agents!

Device and method for extracting voiceprint feature of underwater acoustic target signal

A technology of target signal and voiceprint features, applied in speech analysis, instruments, etc., can solve problems such as limited representation ability of complex functions and restricted generalization ability.

Active Publication Date: 2019-05-24
CSSC SYST ENG RES INST
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] For the voiceprint feature extraction of underwater acoustic detection signals, the limitation of the shallow structure algorithm lies in the limited ability to express complex functions under the condition of limited samples and computing units, and its generalization ability is subject to certain restrictions

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
  • Device and method for extracting voiceprint feature of underwater acoustic target signal
  • Device and method for extracting voiceprint feature of underwater acoustic target signal
  • Device and method for extracting voiceprint feature of underwater acoustic target signal

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0046] The used self-encoding deep neural network of the present invention is not less than three hidden layers, and the node number of each hidden layer is not less than 50, and concrete parameter setting is as shown in table 1, wherein 500 nodes are the frequency of original signal spectrum in the input layer Points, 51 nodes correspond to all frequencies within the value range of the fundamental frequency, and 5 nodes are 5-order harmonic orders from 3 to 7.

[0047] Table 1

[0048]

[0049] Using the above self-encoding deep neural network to extract the voiceprint feature of the underwater acoustic target signal, the obtained results are as follows: Figure 3 ~ Figure 5 shown, where image 3 and Figure 4 Shown is the reconstructed signal spectrum obtained by extracting the fundamental frequency and its harmonics of the target signal through the deep neural network, and reconstructing the original signal spectrum with noise generated by the simulation. imag...

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 hydroacoustic object signal voiceprint feature extraction apparatus and method. The apparatus mainly comprises a signal obtaining module for obtaining an original signal spectrum of hydroacoustic object signals, a feature extraction module for extracting the fundamental frequency and harmonic waves of an object from the original signal spectrum, and a reconstruction module for obtaining voiceprint features by reconstructing original signals by use of the original signal spectrum and the fundamental frequency and harmonic waves of the object. According to the invention, harmonic wave components in the voiceprint features of the hydroacoustic object signals can be effectively extracted, and the apparatus has certain anti-noise interference and anti-frequency-drift capabilities.

Description

technical field [0001] The invention relates to a voiceprint feature extraction device and method, in particular to an underwater acoustic target signal voiceprint feature extraction device and method, belonging to the field of underwater voiceprint feature extraction. Background technique [0002] Ship radiated noise is mainly generated by sound sources such as generators, propulsion systems, and auxiliary equipment on board, and can be detected and acquired by detection equipment. The detected underwater acoustic target signal is different from the characteristic line spectrum of other types of ships corresponding to its multiple sound sources, that is, the voiceprint feature. Voiceprint features include simple features and complex features. The line spectrum in the voiceprint feature is a simple feature. These feature line spectra can be described by frequency, amplitude and width, while the correlation between the feature line spectra is a complex feature, which can be ...

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 Patents(China)
IPC IPC(8): G10L25/18G10L25/30
CPCG10L25/18G10L25/30
Inventor 潘悦吴玺宏李江乔皇甫立
Owner CSSC SYST ENG RES INST
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com