Sequence data identification method and device

A technology of sequence data and recognition methods, applied in the computer field, can solve problems such as long training time, many model parameters, and large data volume requirements, and achieve the effect of reducing data volume and improving integrity and accuracy

Pending Publication Date: 2020-10-02
JINGDONG TECH HLDG CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] In view of this, the embodiments of the present invention provide a method and device for sequence data recognition, which can at least solve the problems of existing model parameters, large data volume requirements, and long training time

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
  • Sequence data identification method and device
  • Sequence data identification method and device
  • Sequence data identification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] Exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present invention to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

[0060] It should be noted that the embodiment of the present invention is mainly applicable to speech-text recognition scenarios, but also has certain versatility for "sequence data" in similar scenarios, but the present invention mainly uses speech-text sequence recognition as an example for illustration.

[0061] To explain sequences here, sequences are objects (or event...

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 sequence data identification method and device, and relates to the technical field of computers. One specific embodiment of the method comprises the following steps: performing sequence segmentation on a voice sequence through a predetermined sequence segmentation mode to obtain a plurality of voice sub-sequences; inputting a plurality of voice sub-sequences into the identification model to obtain elements corresponding to the voice sub-sequences; and under the condition that the number of the elements corresponding to each voice sub-sequence is one, arranging the obtained elements in sequence according to the time sequence of each voice sub-sequence so as to generate a text sequence corresponding to the voice sequence. According to the embodiment, a complex problem is decomposed into a plurality of relatively simple problems, and models for solving the simple problems are connected into an integral solution; because multiple constraints are carried out, the model does not need to mix space and time for modeling, the required parameter quantity and data quantity are greatly reduced, the integrity and accuracy of the text sequence are further improved, anda basis is provided for the interpretability of the whole model.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method and device for identifying sequence data. Background technique [0002] Sequence data recognition is an important consideration in the field of machine learning and artificial intelligence, such as speech recognition technology, which has been widely used in the intelligent interaction of mobile phones, watches, smart home and other devices. With the development of deep learning technology, deep neural network has gradually become the first choice for machine learning with its powerful fitting and generalization capabilities. seq2seq (sequence tosequence model, sequence to sequence model), is a type of end to end (end to end) model, which has achieved great success in sequence data recognition. [0003] Compared with traditional methods, the end-to-end model does not need to extract a large number of features and perform data processing. For example, in speech recogni...

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): G10L15/18G06F40/30G06N3/04G06N3/08
CPCG10L15/18G06N3/08G06N3/045
Inventor 王虎杜强司润泽穆玉芝
Owner JINGDONG TECH HLDG CO LTD
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