Learning Algorithms for Audio Signal Enhancement, Part 2: Implementation of the Rough-Set Method for the Removal of Hiss
Learning algorithms were implemented for the elimination of strong hiss found in old records and of impulse noise affecting transmitted audio signals. The rough-set method was tested with regard to the automatic setting of the cutoff threshold in the spectral filtration of noisy audio. Applied methods, results of musical signal processing, and conclusions are presented.
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 temporarily free for AES members.