This paper examines the use of digital filtering as a method of correcting of performance of audio systems, especially in applications where the desired correction cannot be known in advance, such as room equalization and the restoration of historic recordings. First, the goals of correction are examined in light of differing listener expectations and recording methods. The requirements for objective accuracy in sound reproduction are analyzed, and presently available technology evaluated. Net, analog and digital equalization methods are compared. Analog equalization is shown to be inherently problematic; the frequent need for high-frequency attenuation in sound reinforcement is explained as an aspect of the performance of analog band-pass equalizers. Digital filtering is briefly reviewed; time-domain specification of such filters allows equalization to be applied separately to early and later segments of an elctroacoustic system's impulse response, that is, its direct and reverberant fields. Finally, two digital signal processing methods applicable to equalization are reviewed, the fast Fourier transform method, already extensively applied, and a new system employing adaptive linear prediction and real-time convolution.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.