Combining data from input channels to form output channels that are not linear combinations of the inputs

摘要

Disclosed is a method of extracting N audio output channels from two audio input channels. The method comprises: transforming each of the audio input channels into respective input spectra; creating an inter-channel amplitude spectrum from the input spectra; thresholding each spectral line of the interchannel amplitude spectrum into a single one of N outputs; and combining data from the two input channels according to the spectral mappings to form the N audio output channels. The N audio output channels are not linear combinations of the two input channels. A further method of separating Q audio sources from M audio input channels is disclosed that includes mapping nonlinearly each spectral line of the inter-channel amplitude spectra into a single one of Ne"Q outputs to create a map for each output.