Voice acquisition and identification method for online incrementation

A recognition method and speech technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of inconvenient use, time-consuming, and inconsistent with cognitive habits.

Active Publication Date: 2015-03-25
NANJING UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If there is new input voice data, the probability-related model must be retrained. The training of the model takes a lot of time, and this method does not conform to people's cognitive habits, which makes it very inconvenient to use.

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
  • Voice acquisition and identification method for online incrementation
  • Voice acquisition and identification method for online incrementation
  • Voice acquisition and identification method for online incrementation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] The purpose of the present invention is to develop an automatic and efficient online incremental speech acquisition and recognition method, which will be further described through the embodiments:

[0047] The online incremental speech acquisition and recognition method has the following steps:

[0048] Step 1: First, the terminal for processing voice data receives externally input voice data;

[0049] Step 2: After the terminal for processing voice data receives the externally input voice data, it enters the initialization phase, and the initialization phase is specifically as follows:

[0050] 1) Construct reference point set A={L 1 , L 2}, where the first reference point L 1 , the second reference point L 2 are two random data selected from the externally input speech data;

[0051] 2) Build edge collection Its initial value is an empty set, that is, the first reference point L 1 , the second reference point L 2 There is no initial connection between;

[00...

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

A voice acquisition and identification method for online incrementation includes an initialization stage, an input and competition stage for voice data, a reference point updating stage, a threshold updating stage of an action range and a denoising stage. The defects that a large amount of time is consumed for training a model in the prior art, online incrementation cannot be achieved, this way does not conform to cognitive habits of people and consequently the model is not convenient to use can be effectively overcome.

Description

technical field [0001] The invention belongs to the technical field of speech acquisition and recognition, and in particular relates to an online incremental speech acquisition and recognition method. Background technique [0002] Traditional speech recognition algorithms mostly start from a statistical point of view, and obtain probability-related models by training a large amount of data. If there is new input voice data, the probability-related model must be retrained. The training of the model takes a lot of time, and this method does not conform to people's cognitive habits, which makes it very inconvenient to use. . Contents of the invention [0003] The object of the present invention is to provide an online incremental speech acquisition and recognition method, including an initialization stage, an input and competition stage for speech data, a reference point update stage, a threshold update stage for the scope of action, and a denoising stage. It can effectivel...

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): G10L15/02G10L15/06G10L15/26
Inventor 申富饶许浩然赵金熙
Owner NANJING 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