On the importance of time—a temporal representation of sound
Visual representations of speech signals
Query by humming: musical information retrieval in an audio database
Proceedings of the third ACM international conference on Multimedia
The scientist and engineer's guide to digital signal processing
The scientist and engineer's guide to digital signal processing
Towards a digital library of popular music
Proceedings of the fourth ACM conference on Digital libraries
Music ranking techniques evaluated
ACSC '02 Proceedings of the twenty-fifth Australasian conference on Computer science - Volume 4
Prediction-driven computational auditory scene analysis
Prediction-driven computational auditory scene analysis
Music-listening systems
ICASSP '01 Proceedings of the Acoustics, Speech, and Signal Processing, 2001. on IEEE International Conference - Volume 05
Multipitch estimation and sound separation by the spectral smoothness principle
ICASSP '01 Proceedings of the Acoustics, Speech, and Signal Processing, 2001. on IEEE International Conference - Volume 05
Hi-index | 0.00 |
We present a method for melody detection in polyphonic musical signals based on a model of the human auditory system. First, a set of pitch candidates is obtained for each frame, based on the output of an ear model and periodicity detection using correlograms. Trajectories of the most salient pitches are then constructed. Next, note candidates are obtained by trajectory segmentation (in terms of frequency and pitch salience variations). Too short, low-salience and harmonically-related notes are then eliminated. Finally, the melody is extracted by selecting the most important notes at each time, based on their pitch salience. We tested our method with excerpts from 12 songs encompassing several genres. In the songs where the solo stands out clearly, most of the melody notes were successfully detected. However, for songs where the melody is not that salient, the algorithm was not very accurate. Nevertheless, the followed approach seems promising.