Paper Info
Title
Query by Humming by Using Locality Sensitive Hashing Based on Combination of Pitch and Note
Abstract
Query by humming (QBH) is a technique that is used for content-based music information retrieval. It is a challenging unsolved problem due to humming errors. In this paper a novel retrieval method called note-based locality sensitive hashing (NLSH) is presented and it is combined with pitch-based locality sensitive hashing (PLSH) to screen candidate fragments. The method extracts PLSH and NLSH vectors from the database to construct two indexes. In the phase of retrieval, it automatically extracts vectors similar to the index construction and searches the indexes to obtain a list of candidates. Then recursive alignment (RA) is executed on these surviving candidates. Experiments are conducted on a database of 5,000 MIDI files with the 2010 MIREX-QBH query corpus. The results show by using the combination approach the relatively improvements of mean reciprocal rank are 29.7% (humming from anywhere) and 23.8% (humming from beginning), respectively, compared with the current state-of-the-art method.
Year
DOI
Venue
2012
10.1109/ICMEW.2012.58
ICME Workshops
Keywords
Field
DocType
nlsh vector,extracts vector,locality sensitive,challenging unsolved problem,midi file,content-based music information retrieval,novel retrieval method,candidate fragment,humming error,mirex-qbh query corpus,current state-of-the-art method,music,information retrieval,indexing,midi files,mean reciprocal rank,feature extraction,locality sensitive hashing,vectors,robustness,pitch
Locality-sensitive hashing,Music information retrieval,Pattern recognition,Computer science,MIDI,Speech recognition,Query by humming,Mean reciprocal rank,Hum,Artificial intelligence,Recursion
Conference
ISSN
Citations 
PageRank 
2330-7927
4
0.47
References 
Authors
8
5
Name
Order
Citations
PageRank
Qiang Wang1264.55
Zhiyuan Guo2305.35
Gang Liu3112.01
Jun Guo41579137.24
Yueming Lu512940.73