WSEAS Trans. on Acoustics and Music, vol.1, pp.96-99, 2004 (Refereed Journals of Other Institutions)
In this paper, we develop a method to obtain pitch frequency of monophonic musical signals. The autocorrelation function is used as the main feature to discriminate the notes. Acoustical signals are recorded from an electronic piano, digitized and stored on a computer. Feature extraction, i.e., the short-time autocorrelation computation, is performed and then notes are recognized by using the proposed peak search method. Examples are presented to illustrate the performance of our method.