Numerous techniques support finding sounds that are acoustically similar to a given one. It is hard, however, to find a sound to start the similarity search with. Inspired by systems for image search that allow drawing the shape to be found, we address quick input for audio retrieval. In our system, the user literally sketches a sound effect, placing curved strokes on a canvas. Each of these represents one sound from a collection of basic sounds. The audio feed-back is interactive, as is the continuous update of the list of retrieval results. The retrieval is based on symbol se-quences formed from MFCC data compared with the help of a neural net using an editing distance to allow small temporal changes.
Authors:
Battermann, Michael; Heise, Sebastian; Loviscach, Jörn
Affiliations:
Fachhochschule Bielefeld (University of Applied Sciences), Bielefeld, Germany; Hochschule Bremen (University of Applied Sciences), Bremen, Germany(See document for exact affiliation information.)
AES Convention:
126 (May 2009)
Paper Number:
7794
Publication Date:
May 1, 2009
Subject:
Sound Design and Processing
Click to purchase paper as a non-member or you can login as an AES member to see more options.
No AES members have commented on this paper yet.
To be notified of new comments on this paper you can subscribe to this RSS feed. Forum users should login to see additional options.
If you are not yet an AES member and have something important to say about this paper then we urge you to join the AES today and make your voice heard. You can join online today by clicking here.