Local mean decomposition (LMD) improving method for speech fundamental frequency extraction

A technology of local mean decomposition and pitch frequency, which is used in speech analysis, instruments, etc. to improve calculation accuracy, take into account calculation efficiency and convergence, and improve performance.

Inactive Publication Date: 2019-09-27
XI AN JIAOTONG UNIV
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] At present, there is no report on the iterative envelope method combined with multiple interpolation methods and LMD to obtain the pitch frequency of human speech signals

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
  • Local mean decomposition (LMD) improving method for speech fundamental frequency extraction
  • Local mean decomposition (LMD) improving method for speech fundamental frequency extraction
  • Local mean decomposition (LMD) improving method for speech fundamental frequency extraction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0049] refer to figure 1 , an improved local mean decomposition method for speech pitch frequency extraction, said to use the iterative envelope of the Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) curve PCHIP-IE, said to use the cubic sample The iterative envelope of the bar interpolation (Cubic Spline, CS) curve is CS-IE, including the following steps:

[0050] 1) Given a real adult male speech signal x(t), the sound content is the first letter "A" of the English alphabet, the sampling frequency is set to 48kHz, and the 200ms length signal is intercepted for analysis. The waveform is as follows figure 2 As shown, the counting variable C=1;

[0051] 2) According to C=1, use CS-IE to obtain the upper envelope of speech signal x(t) as and the lower envelope is Such as image 3 As shown, the local mean function m of the speech...

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 local mean decomposition (LMD) improving method for speech fundamental frequency extraction. Firstly, an upper envelope and a lower envelope of a speech signal are calculated through an iterative envelope method of a cubic spline interpolation, and the local mean function and the envelope estimation function of the speech signal are calculated; after screening is completed once, in the subsequence screening process, an iterative envelope method of a piecewise cubic hermite interpolating polynomial is used for calculating the upper envelope and the lower envelope of the speech signal so as to obtain the new local mean function and envelope estimation function; after LMD is completed, the component with maximum energy is selected to calculate the frequency spectrum of the envelope function of the speech signal; and finally, the fundamental frequency of the speech signal is obtained. The iterative envelope methods with an under-envelope elimination function are introduced into an LMD method, a high-order interpolation curve and a low-order interpolation curve are combined to slow down over-envelopes, thus the local mean function and the envelope estimation function of the signal are built, and the calculating efficiency and convergence stability of the LMD method in the fundamental frequency extraction process of the speech signal are improved remarkably.

Description

technical field [0001] The invention belongs to the technical field of local mean value decomposition methods, and in particular relates to an improved local mean value decomposition method for speech pitch frequency extraction. Background technique [0002] Local mean decomposition method (Local Mean Decomposition, LMD) is a multi-component amplitude modulation-frequency modulation (AM-FM) signal decomposition and demodulation method, which can decompose the multi-component into the sum of several single-component AM-FM signals , and the instantaneous frequency and instantaneous amplitude of each component can be obtained, so as to realize the decomposition and demodulation of multi-component AM-FM signals. [0003] The vocal cords are one of the most important vocal organs in the human vocal system. When speaking, the vocal cords are opened by the impact of the subglottic airflow, and then closed again due to the toughness of the vocal cords. The time for each opening and...

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): G10L19/02
CPCG10L19/02
Inventor 张庆贾林山蒋婷婷
Owner XI AN JIAOTONG UNIV
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