Contouring online audio
Existing complaints from online music consumers include information overload, inability to quickly browse through audio and inability to search for similarity among sounds. Therefore the goal of the CUIDADO project lies in developing content-based technologies using MPEG 7 output in order to address these problems. One of the pilot applications that will target these issues is the Online Sound Palette. The Sound Palette involves reusable modules for audio feature extraction with characteristics for innovative retrieval, editing and processing. It can provide access to large centralised sound sample databases either via intranet or over the web. The descriptors of audio material that employ the Mpeg 7 format are key for the elements of the audio material. The descriptors are both textual as well as numerical in format and are processed automatically from the signal. The features that result from this include the ability to automatically classify samples and conduct a similarity search. Additionally the Online Sound Palette can be a tool for online collaborative work because it addresses both audio and music domains. Thus it covers a range of applications from high level music descriptors such as style and rhythm to lower level audio descriptors such as pitch. Consequently, it could meet the information management needs of record labels and copyright societies regarding both content protection and marketing strategies.
