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

Method for recognizing Chinese language whispered pectoriloquy intonation based on acoustic channel parameter

A recognition method and ear speech technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of high recognition difficulty and low recognition rate, and achieve the effect of high recognition rate and significant superiority.

Inactive Publication Date: 2008-10-08
SUZHOU UNIV
View PDF0 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

④Although there is no fundamental tone in ear speech, the tone and pitch can still be perceived by hearing
However, it is difficult to recognize the tone of ear speech by using the amplitude envelope method, and the recognition rate is low.

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
  • Method for recognizing Chinese language whispered pectoriloquy intonation based on acoustic channel parameter
  • Method for recognizing Chinese language whispered pectoriloquy intonation based on acoustic channel parameter
  • Method for recognizing Chinese language whispered pectoriloquy intonation based on acoustic channel parameter

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] Embodiment one: with reference to the Mandarin Chinese tones model proposed by Yang Shun'an, make the four-tone curve of Chinese phonetics, as attached figure 1 As shown, the solid line in the figure is one tone, the short dotted line is two tones, the dotted line is three tones, and the long dotted line is four tones.

[0025] The self-recorded ear speech is used for digital sampling, and the sampling frequency is 8000Hz. Pre-emphasize the voice first, that is, increase the high-frequency part. As a result, the frequency spectrum of the signal is flattened and kept in the entire frequency band from low frequency to high frequency, and the frequency spectrum can be calculated with the same signal-to-noise ratio, so as to facilitate frequency spectrum analysis or channel parameter analysis.

[0026] The pre-emphasis adopts a first-order digital filter: H(z)=1-μz -1 , where H is the transfer function, z is the z transformation, μ is the pre-emphasis coefficient, μ<1.

...

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 present invention discloses a method for recognizing whispering voice tones in Chinese language on the basis of sound track parameters, which comprises: performing digital sampling for the recorded whispering voice, and analyzing the sample data to recognize the whispering voice tone; the method is characterized in: the analysis for the sample data is: performing framing and windowing for the sample data of whispering voice, at a window length no greater than 20ms; calculating linear prediction model parameters of each frame of voice, calculating the gain parameter of each frame of voice signal, and thereby obtaining a gain curve of the voice signal; comparing the gain curve with the reference voice tune curve, to determine the tone of the whispering voice. The present invention employs a sound track gain parameter analyzing method on the basis of sound track parameters to implement recognition of whispering voice tone in Chinese language. The recognition method is applicable to Chinese voice recognition systems, and is can achieve very high recognition ratio and has outstanding advantages.

Description

technical field [0001] The invention relates to a method for speech recognition, in particular to a method for recognizing the tone of Chinese ear speech. Background technique [0002] Otospeech is a pronunciation pattern that differs from normal speech and is characterized by low volume and no vocal cord vibration at all. Whispering, as a special way of language communication, has a wide range of applications. [0003] In medicine, speech clinicians study ear speech patterns to help aphonic patients, and are working to see if whispering can be beneficial for noise recovery and treatment in patients undergoing laryngeal surgery. From a communication point of view, in public places such as conferences, in order to avoid interference to others or to maintain the confidentiality of the conversation, people sometimes need to use whispers for telephone communication. In addition, the research on the topic of ear speech can also provide a basis for the speech recognition and 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
IPC IPC(8): G10L15/00G10L15/02G10L15/08G10L11/04G10L15/12G10L25/90
Inventor 赵鹤鸣龚呈卉
Owner SUZHOU UNIV
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