Method for building voice transformation model and method and system for voice transformation

Abstract

The invention discloses a method for building a voice transformation model and a method and device for achieving voice transformation between first language and second language. The method for transformation includes conducting voice segmentation on first language voice to be transformed to obtain at least one first language syllable, recording syllable duration parameter of each first language syllable and obtained by voice segmentation, extracting base frequency parameter of each first language syllable, determining base frequency parameter and syllable duration of each corresponding second language syllable according to the base frequency parameter and the syllable duration parameter of each first language syllable, adjusting voice waveform of the first corresponding language syllables by the base frequency parameter and the syllable duration according to each second language syllable to obtain voice waveform of each second language syllable and output the syllables. When the method is used for conducting voice transformation, voice quality of input voice and transformed output voice is basically consistent, and real-time transformation can be conducted.

Description

technical field [0001] The invention relates to the technology of realizing different voice conversions by using a computer, in particular to a method for establishing a voice conversion model and a method and system for realizing voice conversion between a first language and a second language. Background technique [0002] Currently, there are two ways to implement the conversion from Mandarin Chinese to dialects. The first commonly used conversion method is complete manual conversion, that is to say, if you want to convert the Mandarin speech into the corresponding dialect speech, you need to know the content of the Mandarin speech, and then read each part of the speech content read out in the dialect. Sentences are recorded. Since there is no limit to the amount of speech content in Mandarin, this results in time-consuming and labor-intensive conversion work. [0003] The second is to use the speech synthesis technology of the computer to realize. A typical Chinese pate...

AI Technical Summary

Problems solved by technology

In this method, at the beginning of the establishment of dialectization rules, it is necessary to establish tagged sentence databases for all dialects and Mandarin, which is equivalent to creating a new text-to-speech system, so it is also time-consuming

[0004] Although the second implementation method has improved adaptability compared with the first implementation method, and can convert any Mandarin text into dialect speech, an obvious problem of the second method is that the dialect speech converted by the speech synthesis system is fixed. The sound quality is very good, and the other is that it can only convert Mandarin text into dialect voice, but cannot realize real-time conversion between Mandarin voice and dialect voice

Images