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

Voice coder

A vocoder and encoder technology, applied in the field of speech signal processing, to achieve the effect of high reconstruction efficiency and strong correlation

Inactive Publication Date: 2017-12-12
TSINGHUA UNIV +1
View PDF5 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

There is no vocoder in the prior art that decomposes and reconstructs speech signals based on content and speakers

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
  • Voice coder

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is only some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0035] figure 1 A schematic structural diagram of a vocoder provided by an embodiment of the present invention is shown, as shown in figure 1 As shown, the vocoder of this embodiment includes: an encoder 1 and a decoder 2;

[0036] The encoder 1 includes: a first acquisition module 11 and a first extraction module 12; wherein:

[0037] The 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 provides a voice coder comprising an encoder and a decoder. The encoder comprises a first acquiring module which is used for acquiring a target voice signal and converting the same into a log domain frequency spectrum, and a first extracting module which is used for extracting content factors of each frame of the target voice signal by using a pre-trained content factor extracting model and extracting speaker factors of each frame of the target voice signal by using a pre-trained speaker factor extracting model and sending the factors to the decoder. The decoder comprises a reconstruction module which is used for reconstructing the content factors into a content spectrum by using a pre-trained content spectrum reconstruction model and reconstructing the speaker factors into a voice print spectrum by using a pre-trained voice print spectrum reconstruction model, and a synthesis model used for adding the content spectrum and the voice print spectrum together to obtain a reconstructed log domain frequency spectrum. The voice coder can decompose and reconstruct voice signals based on content and speakers, thereby being more strongly correlated to tasks and being high in reconstruction efficiency; reconstructed voice signals are relatively close to original voice signals.

Description

technical field [0001] The invention relates to the technical field of speech signal processing, in particular to a vocoder. Background technique [0002] The vocoder analyzes the speech signal at the sending end, extracts the characteristic parameters of the speech signal, encodes and encrypts them to match the channel, and transmits them to the receiving end through the information channel, and the receiving end restores them according to the received characteristic parameters (repeated structure) original voice waveform. [0003] The traditional vocoder is generally based on the source-filter (Source-Filter) model, which decomposes the speech signal x(t) into two parts: the glottal excitation e(t) and the vocal tract modulation h(t), which are written in the convolution form as follows: [0004] x(t)=x(t)×e(t). [0005] This basic decomposition method is the basis of many speech signal processing technologies, such as parametric encoders in speech coding and vocoders in...

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
IPC IPC(8): G10L19/16G10L19/02G10L19/097G10L13/033G10L13/08G10L25/30
CPCG10L13/033G10L13/08G10L19/02G10L19/097G10L19/16G10L25/30
Inventor 王东李蓝天
Owner TSINGHUA UNIV