DocumentCode :
2560799
Title :
Approximate lower-bounding functions for the speedup of DTW for melody recognition
Author :
Lee, Hong-Ru ; Chen, Ching ; Jang, Jyh-Shing Roger
Author_Institution :
Dept. of Comput. Sci., Nat. Tsing Hua Univ., Taiwan
fYear :
2005
fDate :
28-30 May 2005
Firstpage :
178
Lastpage :
181
Abstract :
Conventional methods for melody recognition (or the so-called query by humming) are primarily based on dynamic programming, such as dynamic time warping (DTW). However, the computation load of DTW becomes prohibitively massive when the database has more than, say, 10,000 songs. To lessen the computation while keeping a reasonable recognition rates, this paper present a lower-bound based method that can reduce the computation effectively.
Keywords :
acoustic signal processing; approximation theory; dynamic programming; music; approximate lower-bounding function; dynamic programming; dynamic time warping; melody recognition; query by humming; Autocorrelation; Computational complexity; Computer science; Databases; Dynamic programming; Flowcharts; Information retrieval; Laboratories; Microphones; Speech recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Cellular Neural Networks and Their Applications, 2005 9th International Workshop on
Print_ISBN :
0-7803-9185-3
Type :
conf
DOI :
10.1109/CNNA.2005.1543190
Filename :
1543190
Link To Document :
بازگشت