End-to-end macaque voiceprint verification method and system based on cyclic frame-level feature fusion

A voiceprint verification and feature fusion technology, applied in the computer field, can solve the problems of inaccessibility, difficulty in implementation, and high vigilance of primates, and achieve the effect of expanding limitations and improving accuracy

Active Publication Date: 2021-07-16
NANHAI RES STATION OF INST OF ACOUSTICS CHINESE ACADEMY OF SCI
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Primates mostly live in high mountains and dense forests, it is difficult to effectively observe animals through vision, and primates are very vigilant, and humans are difficult to approach, making it difficult to implement direct observation, DNA fingerprinting and marking methods

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
  • End-to-end macaque voiceprint verification method and system based on cyclic frame-level feature fusion
  • End-to-end macaque voiceprint verification method and system based on cyclic frame-level feature fusion
  • End-to-end macaque voiceprint verification method and system based on cyclic frame-level feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] Such as figure 1 As shown, Embodiment 1 of the present invention proposes an end-to-end rhesus monkey voiceprint verification method based on cyclic frame-level feature fusion, which method includes the following steps:

[0054] Step 110: Select speech pairs from the preprocessed macaque corpus according to the rules, and construct a training set and a test set. Among them, the training set data is randomly divided into q groups, and each group contains the speech segments of m macaques.

[0055] In the prior art, feature extraction is usually performed on speech data in the preprocessing stage to obtain MFCC, LPC, or spectrogram, etc., for classification by classification models. The preprocessing of the present invention is to intercept the effective speech segment, that is, to cut out the silent segment in the original speech, rather than feature extraction, and the macaque corpus after preprocessing is still data in the speech format.

[0056] Step 120: Randomly r...

Embodiment 2

[0107] Such as Figure 4 As shown, Embodiment 2 of the present invention proposes a macaque voiceprint verification system, which is implemented based on the above-mentioned macaque voiceprint verification method. In the training and testing phase, the system specifically includes:

[0108] The data processing module 410 is used to select the speech pair by the macaque corpus after the preprocessing according to the rules, constructs the training set and the test set, and the training set and the test set are grouped;

[0109] The backbone network 420 is used to extract the features of the input rhesus monkey speech segment to obtain the frame level feature vector of the rhesus monkey speech segment;

[0110] The feature fusion network 430 is used to perform cyclic frame interception on the frame-level feature vector output by the backbone network, and then perform feature fusion to obtain the fusion frame feature vector of the macaque speech segment;

[0111] The feature com...

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 an end-to-end macaque voiceprint verification method and system based on cyclic frame-level feature fusion. The method comprises the following steps of preprocessing a to-be-verified macaque voice pair; the macaque voice pair comprising two macaque voice segments; inputting the preprocessed macaque voice pair into a pre-trained macaque voiceprint verification model to obtain a conclusion whether the macaque voice pair to be verified belongs to the same macaque, and realizing voiceprint verification; the macaque voiceprint verification model comprising a backbone network, a feature fusion network and a feature compression network which are connected in sequence; wherein the backbone network is used for carrying out frame-level feature extraction; the feature fusion network is used for performing cyclic frame interception and grouping on the frame-level feature vectors subjected to feature extraction, and mapping the frame-level features into fusion frame features based on a channel weighted fusion mechanism; and the feature compression network is used for compressing the fused frame features to obtain sentence-level features corresponding to the macaque voice segments.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to an end-to-end rhesus monkey voiceprint verification method and system based on cyclic frame-level feature fusion. Background technique [0002] Primates are facing a serious existential crisis. In order to effectively carry out primate protection, it is very important to understand the range of individual animal activities and population changes. And these all rely on individual animal verification and tracking. At the same time, individual animal verification, as a basic research, is an important basis for individual animal tracking and has important research value. [0003] Currently commonly used animal individual verification techniques mainly include manual observation, DNA fingerprinting, marking, image verification and voice verification. Primates mostly live in high mountains and dense forests. It is difficult to effectively observe animals through vision. Primates a...

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): G10L17/26G10L17/06G10L17/02
CPCG10L17/26G10L17/02G10L17/06
Inventor 李松斌唐计刚刘鹏
Owner NANHAI RES STATION OF INST OF ACOUSTICS CHINESE ACADEMY OF SCI
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