Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Voiceprint model training method, voice recognition method, devices, equipment and medium

A model training and model technology, applied in the field of biometrics, can solve the problems of poor speech recognition effect and insufficient data volume of the voiceprint recognition model.

Active Publication Date: 2018-02-09
PING AN TECH (SHENZHEN) CO LTD
View PDF5 Cites 32 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Embodiments of the present invention provide a voiceprint model training method, device, equipment, and medium to solve the problem of insufficient data volume during voiceprint recognition model training
[0005] The embodiment of the present invention also provides a speech recognition method, device, equipment and medium to solve the problem of poor speech recognition effect at present

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 model training method, voice recognition method, devices, equipment and medium
  • Voiceprint model training method, voice recognition method, devices, equipment and medium
  • Voiceprint model training method, voice recognition method, devices, equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] figure 1 A flow chart showing the voiceprint model training method in this embodiment. The voiceprint model training method can be applied to the terminal equipment of financial institutions such as banks, securities, investment and insurance, or other institutions that need to perform voiceprint recognition, so as to use the trained target voiceprint model for voiceprint recognition and achieve artificial intelligence. Purpose. Such as figure 1 As shown, the voiceprint model training method includes the following steps:

[0045] S11: Use training speech data in at least two business scenarios to perform general background model training to obtain a general background model.

[0046]Among them, the business scene refers to the scene where financial institutions such as banks, securities, investment and insurance or other institutions are handling corresponding businesses. The training voice data in a business scenario refers to the voice data collected for training ...

Embodiment 2

[0084] Figure 5 A schematic block diagram of a voiceprint model training device corresponding to a voiceprint model training method in embodiment 1 is shown. Such as Figure 5 As shown, the voiceprint model training device includes a general background model acquisition module 11, a voiceprint feature acquisition module 12, an out-of-domain model acquisition module 13 and a target model acquisition module 14. Among them, the implementation functions of the general background model acquisition module 11, the voiceprint feature acquisition module 12, the out-of-domain model acquisition module 13 and the target model acquisition module 14 correspond to the corresponding steps of the voiceprint model training method in Embodiment 1, in order to avoid repeating , which will not be described in detail in this embodiment.

[0085] The general background model obtaining module 11 is used to carry out general background model training by using the training speech data under at least...

Embodiment 3

[0103] Image 6 A flow chart of a voice recognition method in this embodiment is shown. This speech recognition method is applied in financial institutions such as banks, securities, insurance, or other institutions that need to perform voiceprint recognition, in order to use the trained target voiceprint model for voiceprint recognition to achieve the purpose of artificial intelligence. Such as Image 6 As shown, the speech recognition method includes the following steps:

[0104] S21: adopt the common background model to carry out feature extraction to the voice data to be tested, obtain the voiceprint feature to be tested, and the voiceprint feature to be tested carries the user mark.

[0105] Among them, the voice data to be tested refers to the recording to be recognized, and the general background model is used to extract the features of the voice data to be tested to obtain the voiceprint features to be tested, and the voiceprint features to be tested carry user anno...

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 model training method, a voice recognition method, devices, equipment and a medium. The voiceprint model training method includes the following steps: using training voice data in at least two business scenarios to carry out universal background model training to obtain a universal background model; using the universal background model to carry out feature extraction on intra-domain voice data and out-of-domain voice data to obtain corresponding intra-domain voiceprint features and out-of-domain voiceprint features; carrying out probability linear discriminant analysis model training based on the out-of-domain voiceprint features to obtain an out-of-domain probability linear discriminant analysis model; and adaptively training the out-of-domain probability linear discriminant analysis model based on the intra-domain voiceprint features to obtain a target probability linear discriminant analysis model. The voiceprint model training method overcomes the problem that it is impossible to obtain multiple pieces of voice data of a single person in some business scenarios, and ensures the recognition effect of the target probability linear discriminantanalysis model.

Description

technical field [0001] The present invention relates to the field of biometric identification, in particular to a voiceprint model training method, voice recognition method, device, equipment and medium. Background technique [0002] Voiceprint recognition, also known as speaker recognition, is a type of biometric technology. Voiceprint recognition includes two types of voiceprint recognition technologies: speaker recognition and speaker confirmation. Different types of voiceprint recognition technologies can be used according to different application scenarios. For example, speaker recognition is required in the application scenario of narrowing the scope of criminal investigation. technology, while in the application scenario of bank transaction confirmation, the explainer confirmation technology is required. The so-called voiceprint is the sound wave spectrum that carries speech information displayed by electroacoustic instruments. When speaking, the size and shape of e...

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 Applications(China)
IPC IPC(8): G10L17/04G10L17/14G10L17/02
CPCG10L17/02G10L17/04G10L17/14
Inventor 王健宗郭卉肖京
Owner PING AN TECH (SHENZHEN) CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com