Method for achieving voice navigation robot based on deep learning

A technology of deep learning and voice navigation, applied in the field of telecommunications, can solve the problems of complex unstructured text content, a large number of manual tagging keywords, and the consumption of large human resources, so as to improve labor costs, improve recognition accuracy, and save human effect

Active Publication Date: 2021-01-29
广州云趣信息科技有限公司
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Better results can be obtained by using the traditional rule method, but a lot of manual labor is required to label keywords and maintain the keyword thesaurus in the later stage
However, since 20,000 to 30,000 users use voice navigation robots every day, the unstructured text content becomes complicated, which requires a lot of manual annotation and establishment of a keyword vocabulary, which will consume a lot of human resources. It also requires a lot of time cost

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 achieving voice navigation robot based on deep learning
  • Method for achieving voice navigation robot based on deep learning
  • Method for achieving voice navigation robot based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The present invention will be further described in conjunction with the following application scenarios.

[0032] see figure 1 , which shows a method for implementing a voice-guided robot based on deep learning, including:

[0033] Obtain training corpus: Obtain training call recording text data to form a training corpus, where training call recording text data includes call recording text and corresponding navigation queue labels.

[0034] Data preprocessing: Preprocessing the training call recording text data to obtain standardized training corpus.

[0035] Optionally, after the call recording text is obtained, the text is preprocessed, so as to complete the normalization of the training corpus.

[0036] Optionally, preprocessing the training call recording text data includes the following steps:

[0037] 1) Filter phrases: Filter the sentences whose sentence length is lower than the threshold in the training call recording text.

[0038] 2) Text word segmentation...

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 provides a method for achieving a voice navigation robot based on deep learning. The method comprises the following steps: acquiring call text data of a user; inputting the user call text into a trained navigation queue classification model based on deep learning for prediction and recognition, and obtaining an output navigation queue prediction result; and transferring the user to acorresponding manual queue for processing according to the obtained navigation queue prediction result. According to the method provided by the invention, a large amount of text data can be trained in a targeted manner, and a good effect can be quickly obtained. When new knowledge is encountered in the later period, training can be enhanced to enable the model to adapt to the new knowledge existing in the text data, so that the labor cost and the time cost can be greatly improved.

Description

technical field [0001] The present invention relates to the fields of telecommunication, deep learning and natural language, and in particular to a method for implementing a navigation robot based on deep learning for operator texts. Background technique [0002] At present, the voice navigation robot uses the voice interaction method to identify the user's intention, and then the robot automatically judges the user's situation and docks with the relevant manual queue for processing. This process requires the robot to automatically identify the user's speech information and then transfer to the relevant queue. Better results can be obtained by using traditional rules, but a lot of manual labor is required to label keywords and maintain the keyword thesaurus later. However, since 20,000 to 30,000 users use voice navigation robots every day, the unstructured text content becomes complicated, which requires a lot of manual annotation and establishment of a keyword vocabulary, ...

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): G06F40/279G06F40/289G06N3/04G10L15/26B25J11/00
CPCG06F40/279G06F40/289G10L15/26B25J11/0005G06N3/045
Inventor 黄诗雅罗睦军邓从健
Owner 广州云趣信息科技有限公司
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