Learning algorithms provide a powerful tool when used to eliminate impulse distortions affecting old records. Previously, a neural network simulation was tested with regard to the detection of scratches, hiss, and clicks. New experiments implemented another decision system based on the rough set method. Principles of this relatively new method of considering data are presented. Results of this approach to the sound restoration of old records are shown.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.