Multirate signal processing for multiple listener low frequency room acoustic equalization

Low frequency room acoustic equalization is a challenging problem to solve with realizable low order digital equalization filters. In this paper we address the problem of low frequency room equalization for multiple listeners, simultaneously, using multirate filtering with a pattern recognition approach. The pattern recognition model captures the variabilities in room acoustics between different listening positions, whereas multirate processing allows effective filtering, of low frequency room responses, at low rates. An important feature of the proposed filter is its ability to perform simultaneous equalization of responses at low frequencies at multiple locations with a reduced computational requirement. A comparison of equalization performance between the proposed filter and a lower order filter, designed using a popular warping technique, is also presented.