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

Voiceprint recognition method based on error back propagation neural network

A back-propagation, neural network technology, applied in the fields of pattern recognition, deep learning, and speech signal processing, can solve the problems of voiceprint recognition accuracy and efficiency need to be improved, so as to improve the accuracy and efficiency, reduce the impact of noise, prevent The effect of overfitting

Inactive Publication Date: 2020-08-11
STAR (CHONGQING) INTELLIGENT EQUIP TECH RES INST CO LTD
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of this, the object of the present invention is a voiceprint recognition method based on error backpropagation neural network, to solve the technical problem that the accuracy and efficiency of existing voiceprint recognition still need to be improved

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
  • Voiceprint recognition method based on error back propagation neural network
  • Voiceprint recognition method based on error back propagation neural network
  • Voiceprint recognition method based on error back propagation neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The present invention will be further described below in conjunction with the accompanying drawings and embodiments, so that those skilled in the art can implement it with reference to the description.

[0050] This embodiment is based on the voiceprint recognition method of the error reverse propagation neural network, which includes the following steps:

[0051] Step 1: Collect the speech signal of a known speaker, and preprocess the speech signal after obtaining the speech spectrogram. The preprocessing of the speech signal includes sampling and quantization, pre-emphasis, frame windowing and endpoint detection.

[0052] In order to balance the impact of the attenuation produced by the sound acquisition process on the speech signal, this embodiment uses pre-emphasis to strengthen the high-frequency parts that are easy to lose in the signal other than noise, so that the attenuation in the subsequent transmission can be properly supplemented to make the sound spectrum s...

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 voiceprint recognition method based on an error back propagation neural network, and the method comprises the following steps: step 1, collecting a voice signal of a known speaker, obtaining a voice spectrogram, and then preprocessing the voice signal, and the preprocessing of the voice signal comprises the steps: sampling and quantification, pre-emphasis, framing windowing, and end point detection; step 2, taking a Mel-frequency cepstrum coefficient as a characteristic parameter of the voice signal, and extracting the Mel-frequency cepstrum coefficient of the preprocessed voice signal; step 3, establishing a BP neural network, and training the BP neural network by using a part of the Mel-frequency cepstrum coefficients extracted in the step 1; and step 4, collecting a to-be-recognized voice signal, obtaining a Mel-frequency cepstrum coefficient of the to-be-recognized voice signal according to the method in the step 1, and inputting the Mel-frequency cepstrumcoefficient into the BP neural network trained in the step 2 for recognition to obtain a recognition result. According to the invention, the accuracy and efficiency of voiceprint recognition can be effectively improved.

Description

technical field [0001] The invention relates to the technical fields of deep learning, pattern recognition and voice signal processing, and in particular to a voiceprint recognition method. Background technique [0002] Today, our lives are surrounded by artificial intelligence technology, and the amount of information is huge. The issue of personal information disclosure, such as account numbers, passwords, and personal key information, has become the focus of attention. If information security management issues cannot be effectively guaranteed, personal information will be leaked, which in turn will cause serious economic losses to enterprises and even the country. Verifying the validity of identity information is the primary consideration. With the rapid development of information technology, passwords composed of letters and numbers are not enough to protect personal privacy, and voiceprint recognition is of great significance to protect the privacy of information. Voic...

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): G10L17/02G10L17/18G10L17/22G10L25/24G10L25/30G10L25/45G10L15/05G10L15/16
CPCG10L17/02G10L17/18G10L17/22G10L25/24G10L25/45G10L25/30G10L15/05G10L15/16
Inventor 宋永端陈里蒙
Owner STAR (CHONGQING) INTELLIGENT EQUIP TECH RES INST CO LTD