With the booming of Internet and the rapid growth of digital storage capability, we are exposed to virtually unlimited volume of multimedia information, including movies, video clips, image animation, still images, music, speech, text, etc. Manually searching useful or interesting pieces of information is exactly the same as finding a needle in a haystack. Automatically generating semantically meaningful index for a large volume of documents is critical for efficient and effective multimedia information retrieval. During the last decade, multimedia content index and retrieval, as a new yet fast growing field, has attracted tremendous interest from researchers worldwide. MPEG-7, formally known as Multimedia Content Description Interface, is an internationally collaborated effort that addresses such a challenge. This chapter will focus on the recent progress of audio indexing and retrieval, a major component within the broad scope. Everyday, we are immersed in ubiquitous audio. Very often, we want to search a piece of audio that either we know or heard before, for example, a specific song or a conference recording, or we are not aware of, for example, a piece of symphony or one speech of President Kennedy. With the support of audio indexing and retrieval services, such a task is an enjoyable experience, otherwise, really a headache. In this chapter, we will survey the state of the art in the area of audio content indexing and retrieval, address audio indexing algorithms and query methods, illustrate several representative audio retrieval systems, and briefly introduce MPEG-7 audio standard.
|Original language||English (US)|
|Title of host publication||Handbook of Video Databases|
|Subtitle of host publication||Design and Applications|
|Number of pages||27|
|State||Published - Jan 1 2003|
ASJC Scopus subject areas
- Computer Science(all)