Sound-breath signal separating and synthesizing method and system

Abstract

The present invention discloses a sound-breath signal separating and synthesizing method and system. The method comprises the steps of transforming a sampled voice signal from a time domain to a frequency domain to obtain an original frequency spectrogram; according to the original frequency spectrogram, measuring a fundamental frequency of the voice signal; marking the positions having integer multiple distances with the fundamental frequency as the peak value positions in the original frequency spectrogram, and taking the position between two peaks as a valley value position; extracting the peak value position information and the valley value position information from the original frequency spectrogram respectively to obtain a corresponding sound information frequency spectrogram and a corresponding breath information frequency spectrogram; processing the sound information and the breath information respectively; synthesizing the processed sound information and the processed breath information to obtain the final voice signal. The method and system of the present invention enables the sound information and the breath information to be separated to carry out the subsequent processing, such as the modification, etc., on a sound signal or a breath signal individually, and can be widely used in the signal processing field.

Description

technical field [0001] The invention relates to the field of signal processing, in particular to a method and system for separating and synthesizing acoustic and air signals. Background technique [0002] Speech is the acoustic performance of language, the most natural, effective and convenient means for human beings to exchange information, and it is also a kind of support for human beings to think. Human beings have entered the information age. Using modern means to study speech processing technology enables people to more effectively generate, transmit, store and obtain speech information, which is of great significance for promoting social development. In recent decades, the study of speech has brought scientists and engineers together to form an important discipline: speech signal processing. Speech signal processing technology is referred to as speech processing. It is closely related to linguistics, phonetics, psychology, acoustics, computer science and artificial in...

AI Technical Summary

Problems solved by technology

[0003] However, most of the current speech signal processing technologies for human voice can only extract the sound signal from the vocal cords and oral resonance according to the relationship between the fundamental frequency and the formant, and have not yet been able to extract the air sounds from the lips and teeth from the human voice signal. signal, it is impossible to modify the sound signal or air sound signal separately for subsequent processing

Images