Text-independent speaker identifying device based on line spectrum frequency difference value

Abstract

The embodiment of the invention discloses a text-independent speaker identifying device based on a line spectrum frequency difference value. A method comprises the following steps: the feature extraction step, wherein a line spectrum frequency parameter is converted into the line spectrum frequency parameter difference value through linear conversion, a generation line spectrum frequency characteristic supervector is formed by combining a current frame, a front adjacent frame and a rear adjacent frame; the model training step, wherein distribution of the characteristic supervector is simulated by utilizing a super Dirichlet mixed model, and a parameter in the model is solved; the identifying step, wherein regarding a voice sequence of an identified person, characteristics are abstracted according the step one, then the model obtained in the step two is input, a likelihood value of each probabilistic model is calculated, the largest likelihood value is obtained, and a number of the speaker is confirmed. By means of the method, the text-independent speaker identification rate can be improved, and high practical value is achieved.

Description

technical field [0001] The invention emphatically describes a text-independent speaker recognition system based on linear transformation line spectrum frequency parameters and super Dirichlet mixed model. Background technique [0002] With the development of computer technology, the use of human biometrics (such as fingerprints, voiceprints, and faces) for identification or confirmation has very important research and application values. Speaker recognition is to automatically confirm whether the speaker is in the recorded set of speakers according to the voice parameters reflecting the characteristics of the speaker's physiology and behavior in the voice waveform, and further confirm the identity of the speaker. Speaker recognition includes two parts: speaker identification and speaker confirmation. The speaker identification system usually includes three parts: extracting features that can represent the speaker, training an independent model for each speaker that conforms...

AI Technical Summary

Problems solved by technology

Text-independent speaker recognition technology does not specify the content of the speech, whether it is training or recognition. The recognition object is a free speech signal. It is necessary to find the characteristics and methods that can represent the information of the speaker in the free speech signal. Therefore, it is relatively difficult to build a speaker model

In addition, the text-related identification system is easy to be falsely identified by stealing, and it is inconvenient to use. What the present invention describes is a text-independent identification system

Images