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

Speaker verification method and device and storage medium

A speaker verification and speaker technology, applied in the computer field, can solve problems such as inability to achieve speaker verification and a decrease in the recognition rate.

Pending Publication Date: 2022-08-02
江苏清微智能科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Embodiments of the present disclosure provide a speaker verification method, device, and storage medium to at least solve the problem that the existing voiceprint recognition technology in the prior art cannot process the detailed information of voice features, which leads to a decline in the recognition rate and cannot achieve Technical issues of speaker verification in specific scenarios under the small memory limit of the mobile terminal

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
  • Speaker verification method and device and storage medium
  • Speaker verification method and device and storage medium
  • Speaker verification method and device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] According to this embodiment, an embodiment of a speaker verification method is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings may be executed in a computer system such as a set of computer-executable instructions, and although A logical order is shown in the flowcharts, but in some cases steps shown or described may be performed in an order different from that herein.

[0025] The method embodiments provided in this embodiment may be executed in a mobile terminal, a computer terminal, a server, or a similar computing device. figure 1 A hardware structure block diagram of a computing device for implementing a speaker verification method is shown. like figure 1 As shown, a computing device may include one or more processors (processors may include, but are not limited to, processing means such as a microprocessor MCU or a programmable logic device FPGA, etc.), memory for storing data, and memory for communication fun...

Embodiment 2

[0069] Figure 7 A speaker verification apparatus 700 according to this embodiment is shown, and the apparatus 700 corresponds to the method according to the first aspect of Embodiment 1. refer to Figure 7 As shown, the apparatus 700 includes: a receiving module 710, configured to receive the collected audio data of the speaker from a preset sound collection device, wherein the audio data is the user's specific command utterance, and the audio data is the user's specific command utterance; The first determination module 720 is used to perform data expansion on the audio data to determine the valid audio segment; the second determination module 730 is used to perform feature extraction on the valid audio segment through a preset feature extraction algorithm to determine the audio feature; Three determination modules 740, configured to match the audio features with the voice features of the pre-registered users through the pre-trained voiceprint model, and in the case of match...

Embodiment 3

[0075] Figure 8 A speaker verification apparatus 800 according to this embodiment is shown, and the apparatus 800 corresponds to the method according to the first aspect of Embodiment 1. refer to Figure 8 As shown, the apparatus 800 includes: a processor 810; and a memory 820, connected to the processor 810, for providing the processor 810 with instructions for processing the following processing steps: receiving the collected speaker audio from a preset sound collection device data, in which the audio data is the user's specific command to sound; the audio data is data expanded to determine the effective audio segment; the feature extraction is performed on the effective audio segment through the preset feature extraction algorithm to determine the audio features; and the pre-trained audio The voiceprint model matches the audio features with the voice features of the pre-registered users. When the audio features match the voice features, it is determined that the speaker a...

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 speaker verification method and device and a storage medium. The speaker verification method is used for verifying a speaker at a mobile terminal, and comprises the following steps: receiving collected audio data of the speaker from a preset sound collection device, the audio data being a specific command sound of a user; performing data expansion on the audio data, and determining an effective audio segment; through a preset feature extraction algorithm, feature extraction is carried out on the effective audio segments, and audio features are determined; and through a pre-trained voiceprint model, matching the audio features with sound features of a pre-registered user, and under the condition that the audio features are matched with the sound features, determining that the speaker and the user are the same person, the voiceprint model being a model based on long and short term memory and a global average information fusion technology.

Description

technical field [0001] The present invention relates to the field of computer technology, and in particular, to a speaker verification method, device and storage medium. Background technique [0002] Voiceprint recognition technology is a technology that automatically verifies the speaker's identity based on the comprehensive characteristics of the extracted speaker's voice characteristics and speech content information. Most of the current recognition effects are based on common large models, and the processing of speech features is not detailed enough, which will lead to a decrease in the recognition rate of the model. And it is impossible to realize speaker verification in specific scenarios on the mobile terminal, under the limitation of small memory (within 100k parameters). [0003] In view of the above-mentioned technical problems in the prior art that the existing voiceprint recognition technology cannot process the detailed information of the voice features, result...

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): G10L17/02G10L17/04G06F21/32G06N3/04G06N3/08
CPCG10L17/02G10L17/04G06F21/32G06N3/049G06N3/08
Inventor 胡庆焱王博欧阳鹏
Owner 江苏清微智能科技有限公司