An Approach for Perceptual Similarity Detection Between Audios Independent of Genre Via Metadata Extraction and Correlation;
No Thumbnail Available
Date
2007
Authors
Komsu,F.
Tora, Hakan
Öztoprak,K.
Tora,H.
Journal Title
Journal ISSN
Volume Title
Publisher
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
This study presents an approach for perceptual similarity detection between audios independent of genre. The study is formed of three phases; signal pre-processing as the first phase, metadata extraction via various perceptually compatible features as the second phase, and correlation methodology for similarity identification as the third phase. The performance and relative importance of the selected features for perceptual similarity analysis are presented as testing results. Moreover, relative importance of preprocessing is introduced. Using the proposed methodology, perceptual similarity detection between genre independent audios is achieved with a 96.85% performance. Contribution highly lies on the independency of genre.
Description
Keywords
[No Keyword Available]
Turkish CoHE Thesis Center URL
Fields of Science
Citation
0
WoS Q
Scopus Q
Source
2007 IEEE 15th Signal Processing and Communications Applications, SIU -- 2007 IEEE 15th Signal Processing and Communications Applications, SIU -- 11 June 2007 through 13 June 2007 -- Eskisehir -- 73089