Music voice tone changing method for stabilizing tone quality

Abstract

The invention relates to a music voice tone changing method for stabilizing tone quality, and the method comprises the steps: exporting a spectrum envelope through employing a voice signal, which can be divided into a glottis stimulation component and a channel stimulation response component, and the cepstrum sequence of the voice signal; separating the stimulation component of the voice signal through employing the spectrum envelope; carrying out the processing of the stimulation components of the voice signal through a tone changing algorithm, and changing the pitch of the voice signal; finally enabling the stimulation components to be synthesized again after the spectrum envelope and the pitch are changed, and obtaining a music voice signal with the changed pitch and the stable tone quality. The method provided by the invention is simple, is flexible in implementation, and is higher in practicality.

Description

technical field [0001] The invention relates to the field of speech synthesis, in particular to a music speech tone transposition method with stable timbre. Background technique [0002] In recent years, with the popularization of music entertainment, the sound effect processing of music speech has become a research and application hotspot, and has attracted extensive attention from academia and industry. As we all know, the traditional pitch shift method based on resampling can change the pitch. Its essence is the overall movement of the frequency spectrum of the speech signal on the frequency axis. When the pitch is changed, the spectrum envelope will also change accordingly, resulting in Speech signals lose their original vocal tract response information, such as pitch transitions between male voices, female voices, and children. In order to keep the timbre relatively unchanged, the pitch is changed to achieve effects such as pitch adjustment or harmony. Therefore, how ...

AI Technical Summary

Problems solved by technology

As we all know, the traditional pitch shift method based on resampling can change the pitch. Its essence is the overall movement of the frequency spectrum of the speech signal on the frequency axis. When the pitch is changed, the spectrum envelope will also change accordingly, resulting in The speech signal loses the original vocal tract response information, such as the pitch transition between male voice, female voice and child

Images