Corpus optimization method and device

Abstract

Embodiments of the invention provide a method and a device for corpus optimization. For flaw voice in a first audio, a first sound segment corresponding to the flaw voice is determined from a corpus,and then the first sound segment in the corpus is marked as an illegal sound segment. In a subsequent process of synthesizing audio, candidate sound segments are selected from the legal sound segmentsin the corpus, and then the optimal sound segment is selected from the candidate sound segments, and waveform concatenation is performed on the optimal sound segment, to obtain a synthesized audio. In the process, optimization of the corpus is realized by marking the sound fragments in the corpus, codes are not needed to be corrected, and a optimization process is simple and is low in cost.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of intelligent information processing, and in particular to a method and device for optimizing a corpus. Background technique [0002] Speech synthesis, also known as text to speech (text to speech) technology, is a technology used to convert text information into audible sound information. The speech synthesis method based on waveform splicing is the mainstream speech synthesis method at present. [0003] In the speech synthesis method based on waveform splicing, according to the linguistic features and acoustic parameters, for each text segment of the text to be synthesized, the synthesis algorithm is used to select candidate sound segments from the pre-recorded and marked corpus, and then according to the synthesis algorithm. The target cost and link cost of the candidate sound clips are used to select the optimal sound clip, and then the volume or speech rate of the optimal sound c...

AI Technical Summary

Problems solved by technology

[0004] In the above corpus optimization process, it is necessary to improve the synthesis algorithm, which is complicated and costly

Images