Unlock instant, AI-driven research and patent intelligence for your innovation.

Historical context semantic representation method for multi-round question-answering system

A technology of semantic representation and question answering system, applied in neural learning methods, natural language data processing, instruments, etc., to reduce errors

Pending Publication Date: 2022-02-18
山东新一代信息产业技术研究院有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The main disadvantage of this model is that when there is an error in the data of the ASR or NLU module, the error will be passed to the DST module

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
  • Historical context semantic representation method for multi-round question-answering system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0024] A multi-round question answering system historical context semantic representation method, comprising the following steps:

[0025] S1. The MIC pickup device picks up the sound and transmits the audio data to the ASR module. Through the ASR module, through noise reduction processing, echo cancellation and VAD algorithm processing, the audio data is converted into text data. At the same time, the processed audio is processed through the time series algori...

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 historical context semantic representation method for a multi-round question-answering system, which reduces the influence of an upstream error of a dialogue system on a downstream result, and comprises the following steps of converting voice data into text data through an ASR (Accelerated Solvent Regulation) module, and simultaneously, through a time sequence algorithm, using the processed audio data directly as features of the DST to be input into a DST model; after the NLU module receives the text data, processing the text data through a natural language processing algorithm, and transmitting the feature data to the ASR model; using the DST module for fusing audio feature data input by the ASR module and text feature data input by the NLU module, and carrying out information storage tracking in combination with historical context information.

Description

technical field [0001] The invention relates to a multi-round question answering system historical context semantic representation method, which belongs to the technical field of intelligent question answering. Background technique [0002] In the multi-round question answering system, the DST module is connected to the speech recognition module and connected to the dialog strategy optimization (DPL) module. It is an important module for storing historical context information. It is also the key to a good intelligent dialogue system. [0003] The existing DST module superior mainly accepts data from NLU. The NLU data mainly comes from the speech-to-text conversion of the ASR module. The main disadvantage of this model is that when the data of the ASR or NLU module is wrong, the error will be passed to the DST module. Contents of the invention [0004] The purpose of the present invention is to provide a multi-round question answering system historical context semantic ...

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): G06F40/35G06F40/295G06F40/211G06K9/62G06N3/04G06N3/08
CPCG06F40/35G06F40/211G06F40/295G06N3/08G06N3/045G06F18/241G06F18/253
Inventor 冯卫森冯落落李沛李晓瑜高明王建华尹青山
Owner 山东新一代信息产业技术研究院有限公司