An algorithm capable of extracting multipitch information from a guitar sound is described. The method is based on a two-stage approach. First, the sound signal is segmented in time based on the derivative of the signal envelope. This defines the transients between successive chords. In the second stage, a high resolution FFT is applied to a downsampled version of the signal. This yields a frequency resolution of about 1 Hz, using 1 second time support. An iterative procedure, employing frequency-bins interpolation, is applied to the amplitude spectrum to estimate the possible fundamental frequencies or harmonics. The system has been tested with simulated signals and achieves reliable fundamental frequency detection. With real guitar chords, the performance of the algorithm depends on the harmonic complexity of the sound.
Authors:
Bonnet, Laurent; Lefebvre, Roch
Affiliation:
Research lab on Speech and Audio, University of Sherbrooke, Departement of Electrical Engineering, Canada
AES Convention:
114 (March 2003)
Paper Number:
5772
Publication Date:
March 1, 2003
Subject:
Signal Processing for Audio
Click to purchase paper as a non-member or you can login as an AES member to see more options.
No AES members have commented on this paper yet.
To be notified of new comments on this paper you can subscribe to this RSS feed. Forum users should login to see additional options.
If you are not yet an AES member and have something important to say about this paper then we urge you to join the AES today and make your voice heard. You can join online today by clicking here.