Unlock instant, AI-driven research and patent intelligence for your innovation.

Speech conversion method based on one-to-many codebook mapping

A voice conversion and codebook technology, applied in voice analysis, instruments, etc., to achieve good application prospects, fast voice conversion, and improve the effect of similarity

Inactive Publication Date: 2016-08-17
HOHAI UNIV CHANGZHOU
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the invention is to overcome the deficiencies in the prior art. The voice conversion method based on one-to-many codebook mapping of the present invention can solve the problems of real-time performance and similarity after conversion of the voice conversion system in the actual environment. Reduce the cost of the conversion effect in the process of pursuing real-time voice conversion, thereby improving the similarity between the conversion result and the target voice, and has a good application prospect

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
  • Speech conversion method based on one-to-many codebook mapping
  • Speech conversion method based on one-to-many codebook mapping
  • Speech conversion method based on one-to-many codebook mapping

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0063] Introduce below a specific embodiment according to the speech conversion method of the present invention, specifically as follows,

[0064] Step (1), training phase:

[0065] (A) Source and target human speech are decomposed by a harmonic plus stochastic model to obtain the amplitude and phase values ​​of the pitch frequency trace and harmonic channel spectral parameters. The specific details are described as follows:

[0066] A1) Divide the voice signal into frames, the frame length is 20ms, and the frame overlap interval is 10ms. In each frame, use the autocorrelation method to estimate the fundamental frequency. If the frame is an unvoiced frame, set the fundamental frequency to zero;

[0067] A2) For voiced frames (i.e., frames whose fundamental frequency is not zero), suppose the speech signal s h (n) can be formed by superposition of a series of sine waves:

[0068] s h ( n ) ...

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 voice conversion method based on one-to-many codebook mapping. The one-to-many mapping relationship between a source voice codebook and a target voice codebook is established. A part of voice is extracted randomly from a parallel database, and source and target voice codebooks are rapidly established after sound channel parameters are aligned and extracted. The weights of source and target characteristic parameters in data for training corresponding to respective codebook are estimated, the weight mapping relation between source and target voices is established through the statistics and analysis of the relation between two weights, thus the mapping rule of personality characteristics is grasped, the voice conversion with high quality and fast speed is realized, problems of real-time performance of a voice conversion system and similarity after transformation in an actual environment can be solved, the cost of conversion effect is reduced finally in the process of voice conversion real-time performance pursuit, thus the similarity between a conversion result and target voice is raised, and the method has a good application prospect.

Description

technical field [0001] The invention relates to a voice conversion method based on one-to-many codebook mapping, and belongs to the technical field of voice conversion. Background technique [0002] Speech conversion technology is a technology that takes the voice of someone (called the source) as input, modifies its characteristic parameters, and makes it output a voice with the same semantics but with the voice personality of another speaker (called the target). Simply put, it is to transform the voice of a speaker through some means to make it sound like another speaker's speech. Speech conversion is a relatively new branch in the field of audio signal processing, which belongs to the cross-disciplinary branch , its content not only covers all aspects of speech processing such as speech analysis and synthesis, speaker recognition, speech coding and enhancement, but also involves knowledge in the fields of phonetics, semantics and psychoacoustics. [0003] In recent years...

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 Patents(China)
IPC IPC(8): G10L25/93G10L21/00
Inventor 徐宁胡芳鲍静益刘小峰汤一彬蒋爱民
Owner HOHAI UNIV CHANGZHOU