Speech synthesizing method and device

Abstract

The invention provides a speech synthesizing method and a speech synthesizing device. The speech synthesizing method comprises the steps of: processing text to obtain text to be synthesized; sending the text to be synthesized to an online speech synthesizing system for speech synthesis in the presence of network connection; and if the online speech synthesizing system fails during the process of carrying out speech synthesis or the network connection is interrupted in the actual using process, then sending text on which speech synthesis is not completed by the online speech synthesizing system to an offline speech synthesizing system for speech synthesis. The speech synthesizing method and the speech synthesizing device combines the advantages of online speech synthesis and offline speech synthesis, can provide more stable speech synthesizing services with more natural effect, ensures that the speech synthesizing request of a user can be always completed successfully, and improves the user recognition degree of the speech synthesizing services and the user experience degree.

Description

technical field [0001] The invention relates to the technical field of speech processing, in particular to a speech synthesis method and device. Background technique [0002] Speech synthesis technology can be divided into two types: cloud-based engine-based speech synthesis (hereinafter referred to as "online speech synthesis") and local engine-based speech synthesis (hereinafter referred to as "offline speech synthesis") according to the way the service is provided. Speech synthesis technologies have their own advantages and disadvantages. Online speech synthesis has the advantages of high naturalness, high real-time and no client device resources, but its shortcomings are also very obvious, because the application (Application; hereinafter referred to as: App) using speech synthesis can send a large section of text to The server side, but the voice data synthesized on the server side is sent back to the client that installed the above App in segments, and the volume of v...

AI Technical Summary

Problems solved by technology

Online speech synthesis has the advantages of high naturalness, high real-time and no client device resources, but its shortcomings are also very obvious, because the application (Application; hereinafter referred to as: App) using speech synthesis can send a large section of text to The server side, but the voice data synthesized on the server side is sent back to the client that installed the above App in segments, and the volume of voice data is relatively large even after compression (for example: 4kb / s), if the network environment is unstable , online speech synthesis will become very slow and cannot achieve coherent synthesis; offline speech synthesis can break away from the dependence on the network and can ensure the stability of synthesis services, but the synthesis effect is worse than online synthesis

[0003] To sum up, the products using speech synthesis technology in the prior art are all based on separate online speech synthesis or offline speech synthesis. Online speech synthesis consumes a lot of data traffic, and only prompts the user when a network error occurs. Error, and the effect of offline speech synthesis is not particularly natural, and the user experience is poor

Images