Speaker identification method and system

Abstract

The invention relates to a speaker identification method which comprises the following steps that a neighbourhood preserving embedding space matrix is obtained through training; speaker identification is performed on the basis of the neighbourhood preserving embedding space matrix; the speaker identification based on the neighbourhood preserving embedding space matrix comprises the following steps that principal component analysis (PCA) is performed on a mean super vector X of each gaussian mixture model (GMM), and then a vector W is obtained through dimensionality reduction; the neighbourhood preserving embedding space matrix is used for mapping each vector W, and then a vector W' is obtained; the vector W' obtained by mapping is used as an input feature of a support vector machine (SVM), so as to perform back-end classification modeling; and grading is performed with the help of the SVM, and a speaker is identified in accordance with the grading result. According to the speaker identification method disclosed by the embodiment of the invention, a novel factor analysis technique based on neighborhood preserving embedding (NPE) is adopted, the defects of the existing gross variation factor analysis technique can be effectively overcome, and the speaker identification performance can be further improved.

Description

technical field [0001] The invention belongs to the technical field of speech recognition, in particular, the invention relates to a speaker recognition method and system. Background technique [0002] Speaker recognition technology, in simple terms, is a technology that automatically distinguishes speakers based on voice, so as to identify and authenticate speakers. Speaker recognition has always been of great importance in national security. In addition, with the development of communication and Internet technology, speaker recognition technology has also begun to be applied in multimedia information processing and retrieval. [0003] In the current laboratory environment, because the voice transmission channel is relatively single and the signal-to-noise ratio is high, in this case, the speaker recognition system can achieve good recognition performance. However, in practical applications, the complex and changeable speech environment, such as environmental noise and ch...

AI Technical Summary

Problems solved by technology

However, there are certain deficiencies in the total variation factor analysis technique. On the one hand, the speaker’s annotation information in the training data is not considered in the training process of the total variation space; Can reflect the overall structure of the data

Images