Title :
Pitch Oriented Automatic Singer Identification in Pop Music
Author_Institution :
Dept. of Electr. Eng., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
In this paper, we proposed two novel methods used to distinguish the singer of a pop music. We focused on a single singer and single track case. These two methods are ldquoPitch Extractionrdquo method and ldquo1/12 OFCCrdquo method. The Pitch Extraction method is composed of three stages and they are Singing pitch estimation stage, Exact pitch calculation stage and GMM classification stage. ldquo1/12 OFCCrdquo method is composed of ldquoPitch Feature Calculationrdquo and GMM classification. We also compare these two methods with OFCC based method. With ldquoPitch Extractionrdquo and ldquo1/12 OFCCrdquo method, we have some improvement on works of singer identification using single feature.
Keywords :
music; speaker recognition; automatic singer identification; pitch calculation; pitch extraction; pitch feature calculation; pop music; singing pitch estimation; Cepstrum; Data mining; Hidden Markov models; Human voice; Information filtering; Information filters; Instruments; Interference; Mel frequency cepstral coefficient; USA Councils; GMM; pitch; singer identification;
Conference_Titel :
Semantic Computing, 2009. ICSC '09. IEEE International Conference on
Conference_Location :
Berkeley, CA
Print_ISBN :
978-1-4244-4962-0
Electronic_ISBN :
978-0-7695-3800-6
DOI :
10.1109/ICSC.2009.28