Many-to-many speaker conversion method based on starwgan-gp and x-vector

Abstract

The invention discloses a many-to-many speaker conversion method based on STARWGAN-GP and x-vector, which includes a training phase and a conversion phase, and uses the combination of STARWGAN-GP and x-vector to realize a speech conversion system. This method adds an X-vector vector with better representation performance and practical performance to represent the speaker's personalized features, and uses WGAN-GP to replace GAN, so as to solve the problems of GAN training instability and gradient disappearance, and build a more stable training , a network with a faster convergence speed, further improving the personality similarity and voice quality of the converted voice, and realizing a high-quality voice conversion method. This method can not only relieve the dependence on parallel text and realize speech conversion under the condition of non-parallel text, but also further integrate the conversion system of multiple source-target speaker pairs into one conversion model, that is, realize multi-speaker-to-many Speaker switching.

Description

Technical field[0001]The invention relates to a many-to-many speaker conversion method, in particular to a many-to-many speaker conversion method based on STARWGAN-GP and x vector.Background technique[0002]Speech conversion is a research branch in the field of speech signal processing, which is developed and extended on the basis of speech analysis, recognition and synthesis. The goal of voice conversion is to change the voice personality characteristics of the source speaker so that it has the voice personality characteristics of the target speaker, that is, the voice spoken by one person after conversion sounds like the voice spoken by another person, while retaining the semantics .[0003]After years of research on voice conversion technology, many classic conversion methods have emerged. These include Gaussian Mixed Model (GMM), Recurrent Neural Network (RNN), Deep Neural Networks (Deep Neural Networks, DNN) and most of the speech conversion methods. However, most of these voice c...

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: The technical problem to be solved by the present invention is to provide a multi-to-many speaker conversion method based on STARWGAN-GP and x-vector, which solves the defect that the speaker's individual characteristics are not fully expressed, and overcomes the existing GAN Problems such as unstable training and gradient disappearance not only further effectively improve the personality similarity of the converted voice, but also improve the quality of the converted voice

Images