Man-machine conversation method, electronic equipment and storage medium

Abstract

The invention discloses a man-machine conversation method. The method comprises the following steps: receiving a current statement input by a user; when the semantics of the current statement is incomplete, rewriting the current statement to obtain a new current statement with complete semantics; and determining response content corresponding to the new current statement. The embodiment of the invention has the beneficial effects that the corresponding response content is not urgently determined when the received current statement does not have the complete semantics, but the response content is determined based on the new current statement after the current statement is rewritten to obtain the new current statement with the complete semantics. The defects that errors exist and the accuracy is low when the current statement is directly analyzed under the condition of incomplete semantics are overcome.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a man-machine dialogue method, electronic equipment and a storage medium. Background technique [0002] With the development of voice artificial intelligence, human-machine dialogue services are becoming more and more common in daily life. Users can use natural language to interact with terminal devices and achieve their goals. However, in multi-round human-computer dialogue, since the user's multiple sentences are placed in the entire interaction scene, it is very clear and unambiguous, but if it is separated from the specific scene, a single sentence in the multiple sentences may not have complete semantic information. (It may be due to the existence of omissions, references, etc. in single and). In this case, if only a single statement is used to identify and determine the response content, it will inevitably lead to inaccurate or even wrong answers. Conten...

AI Technical Summary

Problems solved by technology

However, in multi-round human-computer dialogue, since the user's multiple sentences are placed in the entire interaction scene, it is very clear and unambiguous, but if it is separated from the specific scene, a single sentence in the multiple sentences may not have complete semantic information. (It may be due to the existence of omissions, references, etc. in a single and)

In this case, if only a single statement is used to identify and process and determine the response content, it will inevitably lead to inaccurate or even wrong answers.

Images