System and method for playing text voice in intelligent network

Abstract

The present invention discloses a system and a method for playing text speech in an intelligent network. The system comprises a service agent module, a text-to-speech module, a speech agent module and a speech resource module, the speech agent module is responsible for receiving the text, inspecting the text whether converted a speed converting, and finishing the text-to-speech converting throughthe text-to-speech module, performing a numbering to the speech file and returning the speech numbering to a service processing module; the speech resource module is responsible for receiving and storing the speech files sent by the speech agent module, and performing a speech playing responding to the playing request of the service processing module. The present invention makes the same text do not perform the text-to-speech converting on the text-to-speech module, and therefore the performance of speech playing of the system is improved and the onus of the text-to-speech module is reduced. The present invention also has a convenient function of management and data statistics to the text-to-speech process.

Description

technical field [0001] The invention relates to the field of intelligent network, in particular to a system and method for playing text voice in the intelligent network. Background technique [0002] In the current field of voice communication, it is necessary to play various information to users in the form of voice. Among them, playing text information is a very common application. Existing methods such as figure 1 As shown, the basic principle is as follows: [0003] The terminal device requests services from the system through the user access device. After the voice interaction connection is established, when the business processing module needs to play text information, it sends the text information to the text-to-speech module, and the text-to-speech module converts the text into voice. , returned to the business processing module, and then played to the user. [0004] From the above realization principles, the existing method for playing text speech has the follow...

AI Technical Summary

Problems solved by technology

[0005] 1. If the system needs to send the same text message in groups, because each text message needs to be converted into voice by the text-to-speech server, it will greatly affect the system performance

[0006] 2. The resources of the text-to-speech server are limited. If the same text needs to be converted into speech every time, it will be a waste of resources, especially when the business volume is large, which will affect the normal conversion of other texts.

Images