Methods and systems for converting sets of input audio signals in the frequency domain into sets of output audio signals in the frequency domain, and computer-readable, invariant media.

Abstract

This paper discusses methods and systems for converting input audio signals in the frequency domain to output audio signals in the frequency domain, audio filter arrays, filter-based audio systems, and invariant means of reading audio from computer tapes, and multi-input audio processing procedures. Multiple outputs are implemented as linear systems for use in audio filter arrays to convert a frequency-domain input audio signal set into a frequency-domain output audio signal set. The transfer function from an input to an output is defined as a frequency-dependent gain function. According to some implementations, the transfer function includes a direct component, essentially defined as the frequency-dependent gain, and one or more correlated components whose group phase response varies with frequency. The transfer function is generated from a set of sub-band functions, with each sub-band function being generated from a set of corresponding component transfer functions comprising a direct component and one or more decorrelated components.