Single transform perceptual audio encoder

One of the most computationally intensive tasks in a perceptual audio encoder is the time to frequency transformation. The present state-of-the-art encoder, MPEG-AAC, uses a modified discrete cosine transform (MDCT) as its transform engine due to its favorable characteristics. Being a perceptual coder, another crucial module in AAC is the psychoacoustics module, in which the masking threshold is estimated by appropriately considering the effect of each of the masking components. An FFT is performed in this module in order to perform the analysis. The presence of these two transforms has been accepted in MPEG2/4-AAC standard. We explore the possibility of combining the two with the aim of reducing the complexity of the encoder. This technique will also benefit applications with low delay requirement.