MUSIC SYNTHESIS USING PITCH SHIFTING AND ADSR MATCHING

Electronics and Telecommunication Engineering

The sound received at the ears is processed by humans using signal processing that separates the signal along intensity, pitch and timbre dimensions Sound can be represented in time domain as a waveform or in frequency domain as a set of spectra. Along with the pitch analysis, timbre, the vertical dimension of sound drives the research in the field of classification and identification of sounds. The pitch and timbre analysis, hand in hand, also run the manipulation- synthesis processes of musical instruments. And here it becomes the matter of crux whether to highlight pitch analysis or timbre analysis of musical sounds. Timbre, called the color of sound, distinguishes the two different instruments playing the same note. Pitch will be same for these instruments. Timbre analysis involves modeling of spectral characteristics of sound when we think of the transformations, may be time or frequency domain. The spectral parameters of musical sound help the researchers to capture the expressiveness of sound. Considering these aspects, this work helps to select the timbre features across a wide variety and categories and is focused on monophonic sounds. It contributes to study of tumbrel features which are helpful for manipulation of musical notes. Pitch analysis and manipulation talks about the tuning of instruments as well as modifying the human voices. The applications lie with giving sounds to animated characters. Conventional Fourier-based signal processing, while endowed with fast algorithms, is unable to easily represent signal along these attributes Also it’s difficult to process the pitch and timbre separately using these methods. In this paper we use a recently proposed cortical representation (simulation to the processing method of human brain) to represent and manipulate sound. We briefly overview the summary of the process for obtaining, manipulating and inverting cortical representation of a sound and describe algorithms for manipulating signal pitch and timbre separately. The algorithms are first used to create sound of an instrument between a "guitar" and a "trumpet". Applications to creating maximally separable sounds in auditory user interfaces are discussed.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.