Fractal Sinusoidal Modelling For Low Bit-Rate Audio Coding

Abstract

This paper proposes a fractal sinusoidal model that is able to reduce the bit-rate of sinusoidal model coders while achieving perceptually lossless quality. This is achieved by removing the redundancy between sinusoidal tracks through encoding similar tracks with the transformation between a template track and the original track. This paper proposes a transform that is able to capture the perceptual nature of sinusoidal tracks, and can be encoded efficiently. The results from our experiments show that the proposed fractal sinusoidal model coder is able to reduce the bit-rate of the sinusoidal model by roughly 30% while remaining perceptually lossless, while more aggressive modelling results in a reduction of around 60%, with minor quality degradation.