Voiceprint authentication system and method for rapid channel compensation

Abstract

The invention discloses a voiceprint authentication system for rapid channel compensation, which comprises a Gaussian selector, a universal background model (UBM) mixing degree selection module, a speaker factor or channel factor module, a UBM and a speaker model, wherein the Gaussian selector is used for classifying each Gaussian component of the UBM; the UBM mixing degree selection module is used for selecting mixing degree at a relatively shorter distance to calculate zero-order or first-order statistic for each frame of observation vectors of training data by utilizing the Gaussian selector; and the speaker factor or channel factor module is used for estimating a speaker factor or channel factor according to the zero-order or first-order statistic, and establishing the speaker model according to the speaker factor or channel factor. In addition, the invention also discloses a voiceprint authentication method for the rapid channel compensation. By the system and the method, mismatching compensation is realized, and simultaneously a calculated amount is remarkably reduced, thereby remarkably increasing the training speed.

Description

technical field [0001] The invention relates to a voiceprint authentication system, in particular to a voiceprint authentication system with fast channel compensation; in addition, the invention also relates to a voiceprint authentication method with fast channel compensation. Background technique [0002] Voiceprint recognition (Voiceprint recognition), also known as Speaker Recognition (Speaker Recognition), is a technology that automatically identifies the speaker's identity based on the voice parameters in the voice waveform that reflect the speaker's physiological, psychological and behavioral characteristics. Due to the difference in the structure of the congenital vocal tract and the difference in the acquired pronunciation habits, each person's voice is unique. According to this uniqueness, the identity of the person can be accurately distinguished. [0003] In the voiceprint authentication (or voiceprint recognition) system, the mismatch between the training voice a...

AI Technical Summary

Problems solved by technology

[0005] However, the mismatch compensation based on the factor analysis method needs to estimate the EM statistics of all the mixing degrees of the speech parameters of each frame to the UBM. Compared with the feature mapping method, the calculation amount is too large, and the training and testing time is very long, which limits its practical application. application in

Table 1 compares the results of the baseline system and the voiceprint authentication system using factor analysis. The system using factor analysis solves the mismatch between training and test environments and channels, and its false recognition rate is reduced from 8.94% to 3.76%. , but the average time required is increased from 1.26 seconds to 9.53 seconds, which limits the application of factor analysis in practical applications

Images