Text to speech (TTS) method and system

Abstract

The invention relates to a text to speech (TTS) method and system. The method comprises: performing text analysis and linguistic analysis on an inputted text so as to generate front-end scripts comprising corresponding speech units; obtaining, verifying and correcting the front-end scripts; and obtaining the corrected front-end scripts for synthesizing correction speech. By adopting the provided TTS method and system, the errors of the front-end scripts such as word-segmentation errors and polyphone phonetic notation errors can be corrected so that the synthesized speech is more understandable and more user friendly for users, a conventional TTS's shortcomings of insufficient prediction accuracy of rhythm can be overcome, and the synthesized speech is more natural and more expressive.

Description

technical field [0001] The invention belongs to the technical field of speech synthesis, in particular to a speech synthesis method and system. Background technique [0002] The traditional TTS (Text to Speech speech synthesis) system consists of two parts, the front end and the back end. The front end is mainly responsible for text preprocessing and speech unit generation, and the back end is mainly responsible for speech synthesis. Such as figure 1 As shown, both the front end and the back end of the traditional TTS system are set on the client side. The traditional TTS system has many processing links and high computational complexity, which puts forward higher requirements for the computing power and storage capacity of the computer, especially for the emerging mobile Terminal devices such as personal digital assistants, e-books, and mobile phones present great challenges. [0003] Therefore, a distributed TTS system came into being, such as figure 2 As shown, the f...

AI Technical Summary

Problems solved by technology

However, whether it is a traditional TTS system or a distributed TTS system, there is always a problem: with the current artificial intelligence technology, it is impossible to generate absolutely correct front-end results, and there may be deviations in word segmentation, prosody prediction, and phonetic unit selection. The front-end result plays a decisive role in the final synthesis result. A good front-end result greatly improves the intelligibility, naturalness and user acceptance, while a bad front-end result may make the synthesis result far from the text.

Although the common speech synthesis algorithms can synthesize high-quality and natural synthetic speech, they are all based on high-quality front-end analysis results. If there is no high-quality front-end text analysis results as a basis, any speech synthesis No algorithm can synthesize acceptable synthetic speech

Images