Speech processing method and device

Abstract

The invention relates to a voice processing method and device. The method comprises the steps that the voice information is converted into text information when the voice information of the user is received; semantic analysis is performed on the text information so as to obtain multiple operation requests expected to be put forward by the user and the probability of each operation request, whereinthe probability is recorded as the first probability; the current behavior information of the user is acquired and the current behavior information is inputted to a behavior analysis model so as to obtain multiple operation requests expected to be put forward by the user and the probability of each operation request, wherein the probability is recorded as the second probability; the synthesis probability of the operation request is obtained according to the first probability and the second probability of each operation request; and the user is responded according to the synthesis probabilityof each operation request. The actual demand of the user can be reflected so that the interaction frequency between the equipment and the user can be reduced, the user experience can be enhanced and the equipment is enabled to reach the intelligent requirement.

Description

technical field [0001] The present invention relates to the technical field of mobile communication, in particular to a voice processing method and device. Background technique [0002] With the development of speech recognition technology, a large number of applications based on speech recognition technology emerge in an endless stream. The IVR (Interactive Voice Response, Interactive Voice Response) intelligent interactive system based on voice navigation is gradually widely used in various customer service systems. Intelligent service channels are also familiar and accepted by everyone. In order to improve product usability, voice navigation needs to have high intelligent interaction capability, otherwise it will greatly affect the customer's interactive experience and reduce system satisfaction. [0003] At present, the intelligence of the voice navigation system mainly depends on its voice recognition ability and semantic analysis ability. Among them, the speech recog...

AI Technical Summary

Problems solved by technology

[0004] In order to ensure the intelligence of the voice navigation system, the accuracy of the semantic model is very high, and the existing methods to improve the recognition ability of the semantic model mainly include optimizing the model algorithm, or increasing the source of model training data, but it is still difficult to achieve Intelligent requirements

Because the existing voice navigation system is highly dependent on the semantic model, but only relying on the semantic model often cannot reflect the actual needs of users, resulting in the voice navigation system not being able to give users practical feedback, making the voice navigation system unable to meet its intelligent requirements

Images