Title :
Audio Information Retrieval using Semantic Similarity
Author :
Barrington, L. ; Chan, Alvin ; Turnbull, D. ; Lanckriet, Gert
Author_Institution :
Electr. & Commun. Eng., California Univ., La Jolla, CA, USA
Abstract :
We improve upon query-by-example for content-based audio information retrieval by ranking items in a database based on semantic similarity, rather than acoustic similarity, to a query example. The retrieval system is based on semantic concept models that are learned from a training data set containing both audio examples and their text captions. Using the concept models, the audio tracks are mapped into a semantic feature space, where each dimension indicates the strength of the semantic concept. Audio retrieval is then based on ranking the database tracks by their similarity to the query in the semantic space. We experiment with both semantic- and acoustic-based retrieval systems on a sound effects database and show that the semantic-based system improves retrieval both quantitatively and qualitatively.
Keywords :
audio signal processing; information retrieval; acoustic-based retrieval systems; audio tracks; content-based audio information retrieval; semantic feature space mapping; semantic similarity; semantic-based system; training data set; Acoustical engineering; Audio databases; Birds; Content based retrieval; Data engineering; Drives; Image retrieval; Information retrieval; Music information retrieval; Spatial databases; audio retrieval; computer audition; semantic similarity;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
1-4244-0727-3
DOI :
10.1109/ICASSP.2007.366338