Standardized sampling method for extracting pathological speech MFCC features for artificial intelligence analysis

A technology of artificial intelligence and voice, applied in the field of intelligent recognition, to achieve the effect of improving objectivity and efficiency

Pending Publication Date: 2020-08-28
广州科慧健远医疗科技有限公司
View PDF7 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there is no unified method and standard for big data analysis and artificial intelligence research of pathological sp

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
  • Standardized sampling method for extracting pathological speech MFCC features for artificial intelligence analysis
  • Standardized sampling method for extracting pathological speech MFCC features for artificial intelligence analysis
  • Standardized sampling method for extracting pathological speech MFCC features for artificial intelligence analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0053]In speech recognition (Speech Recognition) and voiceprint recognition (Voice Print Recognition), the most commonly used speech feature is Mel-scale Frequency Cepstral Coefficients (MFCC). The human ear has different hearing sensitivities to sound waves of different frequencies. Speech signals from 200Hz to 5000Hz have the greatest impact on speech intelligibility. The critical bandwidth due to sound masking in the low frequency domain is smaller at higher frequencies. Therefore, 28 band-pass filters are arranged from dense to sparse according to the critical bandwidth from low frequency to high frequency to filter the input signal. Taking the signal energy output by each bandpass filter as the basic feature of the signal, this acoustic feature based on the characteristics of the human ear is MFCC. The shape of the human vocal tract can be presented in the form of a short-term power spectrum envelope, and MFCC can accurately represent this envelope, that is, use acousti...

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 standardized sampling method for extracting pathological speech MFCC features for artificial intelligence analysis, and the method comprises the following steps: collecting speech data, carrying out speech data collection on 82 standard Chinese syllables according to the sequence of a standard Chinese speech evaluation system vocabulary, editing the acquired speech data,so that editing work of 82 syllables is finished, then classifying and filing, carrying out signal extraction on the 82 edited syllables, extracting MFCC features of each syllable through specified pre-emphasis, framing, windowing, fast Fourier transform, a triangular band-pass filter and extended framing processing, and forming an MFCC voice library by using the processed data. According to the method, specific MFCC features of each syllable are extracted through a standardized process method, a digitized, standardized and structured speech feature database is constructed, the method can serve pathological speech feature big data and various applications of artificial intelligence analysis, and objectivity and efficiency of pathological speech research and application are improved.

Description

technical field [0001] The invention belongs to the technical field of intelligent recognition, and in particular relates to a standardized sampling method for extracting MFCC features of pathological speech for artificial intelligence analysis. Background technique [0002] At present, the number of people with language barriers in China is increasing year by year, and communication barriers caused by dysarthria seriously affect patients' reintegration into society. Although there are a large number of patients with dysarthria in my country, a survey by Lin Qiang and Lu Jianliang in 2016 found that the current assessment methods cannot meet the needs of therapists for precise speech rehabilitation. Domestic rehabilitation departments and speech rehabilitation institutions still use subjective auditory evaluation and / or scales that require subjective judgment as the main evaluation methods, which lack objectivity and efficiency. In addition, the number of speech therapists ...

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): G10L15/02G10L17/02G10L25/24G10L25/30A61B5/00
CPCG10L15/02G10L17/02G10L25/24G10L25/30A61B5/4803A61B5/725A61B5/7257
Inventor 牟志伟江晨银柯慧明潘正祥温晓宇陈亮朱凌燕
Owner 广州科慧健远医疗科技有限公司
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