Interaction method and system based on virtual human

Abstract

The invention discloses an interaction method and a system based on a virtual human. The method comprises the following steps that multi-modal data of a user is collected and analyzed, a user interaction intention containing the expression intention of the user is obtained, wherein user expression data are extracted from the multi-modal data of the user; the user expression data is analyzed, and the corresponding user expression intention is generated; and corresponding multi-modal interactive response data is generated according to the user interaction intention, the multi-modal interactive response data comprises expression response data, the multi-modal interactive response data are output on the basis of the virtual human, wherein the virtual human is used for displaying the expressioncorresponding to the expression response data. According to the method and the system, the virtual human can show expressions which are naturally matched with current interactive contents, so that acordial feeling of the virtual human is greatly improved, and a user experience is improved.

Description

technical field [0001] The invention relates to the field of computers, in particular to an interaction method and system based on a virtual human. Background technique [0002] With the continuous development of artificial intelligence technology, intelligent robots are more and more used in human production and life. Among the many types of intelligent robots, the more common one is the virtual robot. [0003] In the prior art, a common virtual robot is a virtual image of a human being or a form similar to a human being. This allows virtual robots to simulate human expressions using their avatars. Based on this, in the prior art, a virtual robot is used in many application scenarios to display facial expressions while outputting interaction data. [0004] However, in the prior art, a virtual robot usually outputs a rough expression during human-computer interaction, which is not only very rough, but also has a low degree of matching between the output of the expression ...

AI Technical Summary

Problems solved by technology

[0004] However, in the prior art, virtual robots usually output a rough expression during human-computer interaction. Not only is the output expression very rough, but the matching degree between the output of the expression and the current interactive content is also very low.

In many application scenarios, the expression output of the virtual robot not only fails to bring intimacy to people, but also makes the user feel baffled because it does not match the current interactive content, thus reducing the user experience.

Images