Call mode conversion method based on voice interaction robot

Abstract

According to the conversation mode conversion method based on the voice interaction robot provided by the scheme, the local server can compare the semantic word with the keyword in the corpus database to judge whether the semantic word is stored in the corpus database, if not, the semantic word is added into the corpus database, and if yes, the semantic word is stored in the corpus database; then, the local server judges whether the verbal operation corresponding to the semantic word in the existing corpus database can support continuous communication between the voice interaction robot and the user or not, and if the verbal operation corresponding to the semantic word in the existing corpus database can support continuous communication between the voice interaction robot and the user, the voice interaction robot is started; if yes, controlling a voice interaction device to express verbal skills corresponding to the semantic words; and if not, inquiring whether to switch over the manual seat or not, and if the local server receives a signal for performing manual switch-over within a preset time H, controlling the call to enter a manual switch-over mode by the local server. Through the method, the accuracy of transferring to the manual seat can be improved, and the communication cost of the manual seat is reduced.

Description

technical field [0001] The invention relates to the field of call technology based on a voice interactive robot, in particular to a method for converting a call mode based on a voice interactive robot. Background technique [0002] In customers, hotels, beauty salons, hospitals, buildings, parks, exhibition halls, banks, factories and other scenes with multimedia equipment, service robots are generally installed, not only through service robots to realize voice navigation functions, but also through these Service robots realize voice interaction. For example, users conduct business consultation through voice interaction robots, and voice interaction robots can answer users' questions through voice interaction. The participation of robots greatly saves labor costs. [0003] However, when the existing voice interaction robot talks to the user, it usually uses the preset corpus for the dialogue. When the user triggers a certain keyword and requests to be transferred to a human ...

AI Technical Summary

Problems solved by technology

The trigger keyword setting is relatively limited, which also leads to the rigidity of the existing transfer method, harsh trigger conditions, and relying on a large amount of introductory corpus

In addition, the accuracy rate of the existing transfer to manual agents is relatively low, and the communication cost of artificial agents is also relatively high, and the existing methods cannot be continuously optimized

[0004] Therefore, in view of this situation, it is urgent to design and produce a call mode conversion method based on a voice interactive robot to solve the problem that the accuracy of the transfer of the existing technology is relatively low, the communication cost of the artificial seat is also relatively high, and the existing method cannot The problem of continuous optimization

Images