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

A Parkinson's Speech Enhancement Method Based on Multiband Spectral Subtraction

A technology of speech enhancement and spectral subtraction, applied in the field of Parkinson's speech enhancement based on multi-band spectral subtraction

Active Publication Date: 2020-12-01
NANJING UNIV OF POSTS & TELECOMM
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, speech enhancement methods for the speech of Parkinson's disease patients have not been reported

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
  • A Parkinson's Speech Enhancement Method Based on Multiband Spectral Subtraction
  • A Parkinson's Speech Enhancement Method Based on Multiband Spectral Subtraction
  • A Parkinson's Speech Enhancement Method Based on Multiband Spectral Subtraction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0087] A method for Parkinsonian speech enhancement with multiband spectral subtraction, such as figure 1 Shown: Include the following steps:

[0088] S1: Collect voice signals from Parkinson's patients in a non-contact manner, and collect voice signals for a period of time in a quiet indoor environment. The pure speech signal is modeled as x(t), the noise is modeled as d(t), and d(t) obeys a zero-mean Gaussian distribution, then the collected noisy speech of Parkinson's disease patients is:

[0089] y(t)=x(t)+d(t)

[0090] S2: Sampling and dividing the speech of Parkinson's disease patients collected in step S1 into short-time frames; wherein, sampling includes sampling frequency, sampling precision and noise-containing speech signal after sampling, and the sampling frequency is set to F s kHz, the sampling accuracy is Bbit, and the noise-containing speech signal after sampling is y(m), where m is the discrete time measurement unit; the frame length divided by the short-tim...

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 Parkinson's disease speech enhancement method based on multi-band spectral subtraction. The Parkinson's disease speech enhancement method comprises the following steps: carrying out speech signal acquisition on a Parkinson's disease patient through a non-contact type manner; carrying out sampling and short time frame division on an acquired speech of the Parkinson's disease patient; carrying out short time energy calculation on each frame of the speech, and carrying out self-adaptive frequency band division according to short time energy; calculating a short time power spectrum, smooth parameters of time variation and a smooth power spectrum of the speech containing noises in each window; calculating a deviation correction factor; searching a minimum value of power spectrum density; calculating and updating the power spectrum density of the noises; carrying out speech enhancement by adopting the multi-band spectral subtraction. According to the Parkinson's disease speech enhancement method disclosed by the invention, a recorded speech of the Parkinson's disease patient are subjected to effective noise elimination treatment under a non-experiment environment, so that the speech quality of the Parkinson's disease patient is improved; finally, the accuracy of diagnosing Parkinson's disease or predicating the severe degree of the Parkinson's disease is effectively improved.

Description

technical field [0001] The invention relates to a method for enhancing speech, in particular to a method for enhancing Parkinson's speech based on multi-band spectrum subtraction, and belongs to the field of speech enhancement. Background technique [0002] Parkinson's disease is mainly caused by the variation of dopaminergic neurons in the substantia nigra compacta of the midbrain and the decline in the biosynthesis of dopamine in the residual neurons, resulting in a gradual lack of dopamine in the striatum. At present, the reason for the reduction of dopamine transmitters in the striatum is still unknown, which is also an important reason why Parkinson's disease can only be treated with rehabilitation, but cannot be completely cured. Parkinson's disease will bring great inconvenience to the elderly population and their family life. Especially in patients with advanced Parkinson's disease, they usually have symptoms of muscle stiffness and can only stay in bed for a long t...

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): G10L25/66G10L25/21G10L25/18G10L25/45G10L21/0232G10L21/0216
CPCG10L21/0216G10L21/0232G10L25/18G10L25/21G10L25/45G10L25/66
Inventor 季薇林钢李云
Owner NANJING UNIV OF POSTS & TELECOMM
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com