Natural synthesis of North Indian musical instruments

Abstract

In this endeavor, it is attempted to generate an automatic audio output from a given text file of musical notations of North Indian Classical Music via synthesizing of different musical instruments. The mathematical model developed in this attempt is based on the fundamental frequency component together with extracted envelop of one specified note from the computer acquired waveform to produce more natural music. The model is validated by establishing the fact that the mean square error between the actual waveform and produced by the mathematical model lie within accepted range in L1 and L∞. Among the methods used for synthesis, always there have been concerns for naturalness of music created by mathematical modeling. The limitations associated with this approach are also discussed together with examples.