Method and device for singing imitation

A model, variance technology, used in speech analysis, speech recognition, instrumentation, etc.

Active Publication Date: 2017-09-05
福建凯米网络科技有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] For this reason, it is necessary to provide a method of singing imitation to solve the problem of converting the timbre of the source singer to imitate the timbre of the target singer

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
  • Method and device for singing imitation
  • Method and device for singing imitation
  • Method and device for singing imitation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0065] In order to describe in detail the technical content, structural features, achieved objectives and effects of the technical solution, the following detailed descriptions are given in conjunction with specific embodiments and accompanying drawings.

[0066] The method of singing imitation described in the first embodiment of the present invention is used to convert the voice of the source singer into a voice with the timbre of the target singer. The method includes a training phase and a conversion phase, where:

[0067] For the training phase, please refer to figure 1 , Including steps S101-S105:

[0068] S101: Use the STRAIGHT model to calculate the fundamental frequency sequence value and the spectral envelope coefficient of the audio material corresponding to the source singer and the target singer respectively;

[0069] Adaptive Weighted Spectral Interpolation (STRAIGHT) is a high-quality speech synthesizer that can effectively separate the speech source and filter, decompo...

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 present invention proposes a method of singing imitation, including: preparing audio materials corresponding to the source singer and the target singer; using the STRAIGHT model to analyze the voice characteristics of the source singer and the target singer; using the Gaussian mixture model to train and obtain the source singer A joint GMM model with the target singer; a Gaussian transform function based on inter-frame correlation is used in the timbre conversion process; the timbre of the source singer is mixed proportionally with the timbre of the target singer; the STRAIGHT model is used to reconstruct the sound with the timbre of the target singer; The present invention also provides a device for realizing the above method. The present invention can convert the voice of the source singer into the sound with the timbre of the target singer, the converted sound quality is good, the timbre is close to the target singer, and the timbre characteristics of the target singer can be added in proportion; especially, when the target singer is a singer When it is used, it can greatly improve the confidence and interest of users in digital audio-visual venues, and improve the level of users' imitation of singers' timbre.

Description

Technical field [0001] The invention relates to a method of timbre conversion, in particular to a method of singing imitation. Background technique [0002] With the popularization of smart terminals, people have higher and higher requirements for intelligent life services, and voice intelligent services have become an urgent need for people. [0003] The existing singing compensation system has the technology of correcting the singer's pitch, so that the singer can still sing according to the standard pitch when the singer is out of tune, but less involves the technology of converting the singer's timbre to imitate the singer's timbre. . The intelligence of the K song system urgently needs a technology that allows users to imitate the singing of singers through machine assistance, and the timbre is close to that of professional singers, thereby greatly improving the self-confidence and interest of users in singing, and improving the level of users to imitate the timbre of singers...

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): G10L15/06G10L13/02
Inventor 王子亮邹应双
Owner 福建凯米网络科技有限公司
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