Continuous Pitch Contour as an Improvement Feature for Music Information Retrieval by Humming/Singing

Authors:
Tri Nguyen Truong Duc;Minh Nhat;Ha Nguyen Duc Hoang;Quan Vu Hai
Affiliations:
Faculty of Information Technology, University of Natural Sciences VNU-HCMC, Vietnam;Faculty of Information Technology, University of Natural Sciences VNU-HCMC, Vietnam;Faculty of Information Technology, University of Natural Sciences VNU-HCMC, Vietnam;Faculty of Information Technology, University of Natural Sciences VNU-HCMC, Vietnam
Venue:
PRICAI '08 Proceedings of the 10th Pacific Rim International Conference on Artificial Intelligence: Trends in Artificial Intelligence
Year:
2008

Citing 2
Cited 0

Query by humming: musical information retrieval in an audio database

Proceedings of the third ACM international conference on Multimedia
Name that tune: a pilot study in finding a melody from a sung query

Journal of the American Society for Information Science and Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present a method for smoothing the pitch sequence to remove the outliers caused by pitch tracking method and errors in sung/hummed voice. This approach is used for constructing a query by singing/humming system. After the pitch sequence is smoothed, the continuous pitch contour is calculated. Then, this feature can be used with Dynamic Time Warping (DTW) matching to compute the difference score between the each query and song. Experimental result of this method on TCS Corpus for query by Singing/Humming will be reported and discussed in detail in this paper.