Supercharge Your Innovation With Domain-Expert AI Agents!

Realization method based on improved DTW (dynamic time warping) speech recognition algorithm

A technology of dynamic time warping and speech recognition, which is applied in speech recognition, speech analysis, instruments, etc. It can solve the problems that affect the recognition efficiency of the dynamic time warping algorithm, consume a lot of memory, and take a long time to calculate, so as to improve the calculation speed and recognition efficiency , improve the speed of calculation, reduce the effect of calculation

Inactive Publication Date: 2018-07-24
SOUTHEAST UNIV WUXI INST OF TECH INTEGRATED CIRCUITS
View PDF9 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] It can be seen that the above-mentioned traditional dynamic time warping algorithm is realized by calculating the cumulative distance of a large number of matching paths, and there are still technical problems such as excessive calculation, excessive memory usage, and long calculation time, which greatly affect the dynamic time. Recognition Efficiency of Regular Algorithm in Speech Recognition

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
  • Realization method based on improved DTW (dynamic time warping) speech recognition algorithm
  • Realization method based on improved DTW (dynamic time warping) speech recognition algorithm
  • Realization method based on improved DTW (dynamic time warping) speech recognition algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The present invention will be further described below in conjunction with the accompanying drawings.

[0034] like figure 1 Shown is an implementation method based on an improved dynamic time warping speech recognition algorithm. According to the characteristics of boundary conditions, continuity and monotonicity constraints when selecting a matching path from a reference template and a test template, the selection range of the two matching paths is limited. In the parallelogram area close to the diagonal on the similarity matrix, that is, the constraint of the boundary range is further added in the matching process.

[0035] In this embodiment, the following steps are specifically included:

[0036] S1: Speech sequence Q=[q that passes the test template 1 , q 2 ,...,q i ,...,q n ] (where, n=N is the total number of frames of the speech sequence of the test template, q i is the feature value of each frame in the speech sequence) and the speech sequence of the refe...

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 realization method based on an improved DTW (dynamic time warping) speech recognition algorithm. The method is characterized by, according to the characteristics of meeting boundary conditions, continuity and monotonic constraints when a reference template and a test template select the matching path, limiting the selection range of the matching path of the two within a parallelogram range close to diagonals on a similarity matrix, so that matching distance of time frames corresponding to path points outside the parallelogram does not need to be calculated, and all frame matching distance and cumulative distance matrix do not need to be kept. For the problem of large calculation burden due to too many matching paths in the matching process of a conventional DTW algorithm, the method, by further adding constraints on the boundary range into the path matching process, successfully filters out some paths irrelevant with the template matching final result, therebygreatly reducing unnecessary calculation and memory usage in the matching process and effectively improving calculation speed and identification efficiency of the DTW (dynamic time warping) speech recognition algorithm.

Description

technical field [0001] The invention relates to an implementation method based on an improved dynamic time warping speech recognition algorithm, belongs to the technical field of speech recognition control, and can be used in the technical field of embedded speech recognition which is sensitive to calculation amount and memory usage. Background technique [0002] With the progress of human society and the rapid development of science and technology, people have begun to pursue a smart and convenient home environment. The application of voice recognition control technology in smart homes has become particularly important. It can free people from the trouble of manually controlling equipment. The purpose of controlling home appliances can be achieved through voice, so voice recognition control has become a popular research direction. The development of speech recognition technology has been relatively mature on the PC (computer) device with more available resources and strong ...

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): G10L15/28
CPCG10L15/285
Inventor 刘昊吕修任姚国良
Owner SOUTHEAST UNIV WUXI INST OF TECH INTEGRATED CIRCUITS
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com