Sound track spectrum Gaussian mixture model based rapid voice conversion system and method

A Gaussian mixture, fast speech technology, applied in speech synthesis, speech analysis, instruments, etc., can solve problems such as large amount of calculation, long operation time, inaccurate calculation results, etc., to improve system performance, strong correlation and overlap. Effect

Inactive Publication Date: 2015-03-04
CHANGZHOU INST OF TECH
View PDF6 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Each process involves complex signal processing calculations, which require high software and hardware configuration, and the calculation time is relatively long, which is not conducive to the instantiation of voice conversion technology on some mobile devices and embedded devices with a wide range of applications.
Especially in the stage of feature parameter extraction, the traditional speech conversion system often needs the transformation between time domain, frequency domain and cepstrum domain, and the calculation amount is extremely huge.
In addition, limited by specific hardware devices, overly complex parameter extraction algorithms will also lead to inaccurate calculation results

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
  • Sound track spectrum Gaussian mixture model based rapid voice conversion system and method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention will be further described below in conjunction with the accompanying drawings.

[0036] Such as figure 1 As shown, a fast speech conversion method based on vocal tract spectral Gaussian mixture modeling, is characterized in that the steps include feature parameter extraction and synthesis, feature parameter time alignment, feature parameter training and conversion;

[0037] The feature parameter extraction is to decompose the original speech signal, and the feature parameter synthesis is the reverse process of feature parameter extraction;

[0038] The time alignment of the characteristic parameters is used to arrange and screen the characteristic parameters of the voices of the converted object and the converted object, so as to obtain a set of characteristic parameters synchronized in time;

[0039] The feature parameter training is used to learn the mapping relationship between the conversion object and the voice feature parameter set of the co...

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 sound track spectrum Gaussian mixture model based rapid voice conversion system and method. The method comprises the steps of parameter extraction and synthesis, characteristic parameter time aligning and characteristic parameter training and conversion. By the technologies of fixation of the Gaussian average on Mel frequency spectra, adaptive Gaussian variance adjusting, selecting of sampling points as weight coefficients on logarithm magnitude spectra and the like, the calculation complexity of voice parameter characterization is greatly reduced, and the operating rate is improved greatly.

Description

technical field [0001] The invention relates to a speech signal processing technology, in particular to a fast speech conversion system and method based on vocal tract spectral Gaussian mixture modeling. Background technique [0002] In order to realize the task of speech conversion, it needs to be completed in several steps: feature parameter extraction, parameter matching, mapping relationship construction, parameter real-time conversion, etc. Each process involves complex signal processing calculations, which require high software and hardware configuration, and the calculation time is relatively long, which is not conducive to the instantiation of voice conversion technology on some mobile devices and embedded devices with a wide range of applications. Especially in the stage of feature parameter extraction, the traditional speech conversion system often needs the transformation between time domain, frequency domain and cepstrum domain, and the calculation amount is extr...

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): G10L13/08G10L13/00
Inventor 鲍静益徐宁
Owner CHANGZHOU INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap