High-accuracy semantic comprehension identification method based on word slot sequence model

A semantic understanding and high-accuracy technology, applied in speech recognition, natural language data processing, speech analysis, etc., can solve problems such as difficult configuration, inapplicability, and inflexibility, avoiding the need for a large amount of training data, solving inflexibility and inflexibility Effects that are not easy to configure

Pending Publication Date: 2020-12-29
成都小美伴旅信息技术有限公司
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Voice interaction requires machines to be able to understand colloquial sentences. The current mainstream models are the rule template model and the neural network model. For the rule template model, it has disadvantages such as infle

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
  • High-accuracy semantic comprehension identification method based on word slot sequence model
  • High-accuracy semantic comprehension identification method based on word slot sequence model
  • High-accuracy semantic comprehension identification method based on word slot sequence model

Examples

Experimental program
Comparison scheme
Effect test

specific example

[0031] The short sentence entered by the user "Ah that plays Jay Chou's music Daoxiang", its word slot sequence "#play-@PER -#music-.+ " has the highest matching score, and its specific processing is as follows:

[0032] (1) Use the enumerable word slot "#play" for content positioning, the positioning subscript starts counting from 0, and the starting position of the key information corresponding to the word slot is 3, and then Delete the content before the position subscript 3 in the input sentence, and then use the regular expression to match the content, the word "play" can be extracted, and finally the remaining content "Jay Chou's Music Daoxiang" is given to Next word slot processing;

[0033] (2) Use the part-of-speech type word slot "@PER "Content positioning, wherein PER is the part of speech, the word slot works according to the part of speech, and it is found that the part of speech of "Jay Chou" in the input content "Jay Chou's Music Daoxiang" from the prev...

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 high-accuracy semantic comprehension identification method based on a word slot sequence model, and the method comprises the following steps: formulating a plurality of wordslot matching rules, extracting key information from a received spoken language word sequence by using the plurality of word slot matching rules, and finally carrying out semantic comprehension and dialogue intention identification by using the extracted key information. According to the method, the system has a semantic comprehension capability, a voice interaction function can be further provided, the problems that a rule template model is inflexible and not easy to configure are solved, and the problem that a neural network model needs a large amount of training data is avoided.

Description

technical field [0001] The invention relates to the field of voice interaction, in particular to a high-accuracy semantic understanding and recognition method based on a slot sequence model. Background technique [0002] Information systems or control systems are increasingly using voice interfaces to interact quickly and directly with users. As the functions of these systems are becoming more and more complex and the required dialogue methods are becoming richer, people are Enter the field of large-vocabulary spoken continuous speech recognition. Voice interaction requires machines to be able to understand colloquial sentences. The current mainstream models are the rule template model and the neural network model. For the rule template model, it has disadvantages such as inflexibility and difficult configuration; for the neural network model, it needs A large amount of training data is used as a starting condition, which is not suitable for fields that do not currently hav...

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): G06F40/35G06F40/289G10L15/26
CPCG10L15/26G06F40/289G06F40/35
Inventor 苏生陈睿
Owner 成都小美伴旅信息技术有限公司
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