S. Pfeiffer and J. Robert-Ribes and D. Kim, “Audio Content Extraction from MPEG-encoded sequences”, First International Workshop on Intelligent Multimedia Computing and Networking, IMMCN 2000, Fifth Joint Conference on Information Sciences, JCIS 2000 February 2000 Atlantic City, NJ, USA.

Abstract

Audio content analysis is nowadays mainly performed on quantised sound waves after transforming the samples into the frequency domain. Our work exploits MPEG-1 encoded audio data for audio content analysis. As MPEG-1 uses Subband Coding to compress sound samples, the encoded data is directly usable for content analysis. This has the advantage of real-time content access without having to create a separate index. We discuss the types of content information that can be extracted from the encoded data. We also give experimental results of the quality of the extractable content using silence detection for sound segmentation.