Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Information processing method and device, electronic equipment, and storage medium

An information processing method and memory technology, applied in the information field, can solve problems such as sound distortion, discontinuity, and weirdness, and achieve the effects of avoiding poor effects, improving effects, and improving effects

Active Publication Date: 2018-06-22
MIGU CO LTD +1
View PDF12 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the many nonlinear components of speech, the conversion effect is not good.
In the prior art, a speech conversion method based on a Gaussian mixture model is proposed, which greatly improves the effect of speech conversion, but there are problems of overfitting or discontinuity in the Gaussian mixture model, resulting in the converted voice Compared with the voice of a real person, there is a large distortion, which leads to the conversion is still very weird, so the effect of voice conversion in the existing technology still needs to be further improved

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
  • Information processing method and device, electronic equipment, and storage medium
  • Information processing method and device, electronic equipment, and storage medium
  • Information processing method and device, electronic equipment, and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057] The technical solutions of the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0058] like figure 1 As shown, this embodiment provides an information processing method, including:

[0059] Step S110: extracting the first frequency domain feature of the source speech;

[0060] Step S120: extracting the second frequency domain feature of the target speech;

[0061] Step S130: Construct a Gaussian model based on the first frequency domain feature and the second frequency domain feature;

[0062] Step S140: Map the first frequency domain feature and the second frequency domain feature located in the first space to a second space through nonlinear mapping, wherein the dimension of the second space is higher than that of the first dimension of space;

[0063] Step S150: decomposing the frequency domain features mapped to the second space into a kernel non-negative matrix to obtain a firs...

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 an information processing method and device, electronic equipment and storage medium. The information processing method includes: extracting a first frequency domain feature ofsource speech; extracting a second frequency domain feature of target speech; constructing a Gaussian model based on the first frequency domain feature and the second frequency domain feature; mapping the first frequency domain feature and the second frequency domain feature located in a first space to a second space by a non-linear mapping, wherein the dimension of the second space is higher than the dimension of the first space; performing a KNMF on the frequency domain characteristics mapped to the second space to obtain a first conversion function; and mixing the first conversion functionand the Gaussian model to obtain a second conversion function, wherein the second conversion function is used for converting the sound parameter of the source voice to the sound characteristic parameter of the target voice so as to improve the conversion effect of the voice.

Description

technical field [0001] The present invention relates to the field of information technology, in particular to an information processing method and device, electronic equipment and a storage medium. Background technique [0002] Speech conversion is to convert one person's voice into another person's voice under the condition of preserving the original semantics, so as to realize the voice replacement under the condition of preserving the semantics. [0003] Speech conversion has been widely used in application scenarios such as speech enhancement, speech assistance, and secure communication. [0004] Existing voice conversion includes: using vector quantization spectrum mapping method, multiple linear regression and other methods for voice conversion. However, due to the many non-linear components of speech, the conversion effect is not good. In the prior art, a speech conversion method based on a Gaussian mixture model is proposed, which greatly improves the effect of spe...

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): G10L21/007G10L25/24G10L21/013
CPCG10L21/007G10L21/013G10L25/24G10L2021/0135
Inventor 徐嵚嵛李琳周冰
Owner MIGU CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products