Search Sounds: An audio crawler focused on weblogs


In this paper we present a focused audio crawler that mines audio weblogs (MP3 blogs). This source of semi-structured information contains links to audio files, plus some textual information that is referring to the media file. A retrieval system —that exploits the mined data— fetches relevant audio files related to user’s text query. Based on these results, the user can navigate and discover new music by means of content-based audio similarity. The system is available at:

Extracted Key Phrases


Citations per Year

52 Citations

Semantic Scholar estimates that this publication has 52 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Celma2006SearchSA, title={Search Sounds: An audio crawler focused on weblogs}, author={{\`O}scar Celma and Pedro Cano and Perfecto Herrera}, booktitle={ISMIR}, year={2006} }