Arbitrarily adjustable method and device for changing phoneme naturally

Abstract

The invention discloses an arbitrarily adjustable method and device for changing phoneme naturally. The method comprises the steps of: modeling a voice signal in a sound channel; then obtaining a pronouncing system model, that is, a model of the sound channel; after that, changing the model of the sound channel; and finally restoring the voice signal in a new sound channel model, wherein the newly output voice signal is a signal after the phoneme change. The device comprises a power source model, an input and output module, a control module and a processing module, wherein the control module sends a parameter to the processing module and controls the system to start and stop operating; the input and output module inputs a sound signal, converts the sound signal to a digital signal and sends the digital signal to the processing module; the processing module processes the signal based on the input parameter to form a phoneme changing signal, then sends the phoneme changing signal to the input and output module and then converts the signal to an analogue signal to be output by a sound box. According to the invention, the signal after a phoneme change can have natural sound effect and can be adjusted arbitrarily. The device disclosed by the invention has the advantages of being simple, easy to realize, low in equipment cost and convenient to use.

Description

Technical field [0001] The present invention relates to speech signal processing technology, in particular to a method and device for modeling a speech signal in a sound channel, and then changing the model of the sound channel to obtain a voice-changing signal. Background technique [0002] Using signal processing methods to change the pitch and timbre of the voice signal, to achieve voice change, and even the mutual conversion of male and female voices, is not only an important theoretical research content of voice signal processing, but also has strong application value. The frequency range of male and female voices is very different. The frequency range of male voice is about 80-7000HZ, while the frequency range of female voice is about 150-11000HZ, which is almost twice as high as that of male voice. If you want to achieve the mutual conversion of male and female voices, you must change the frequency of the voice signal to increase or decrease a level, so that the purpose of...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the disadvantages of the existing voice-changing methods, such as unnatural, machine-smelling, inability to change into any human voice at will, and provide a method and device for natural voice-changing that can be adjusted arbitrarily. The specific technical scheme is as follows

Images