The application of speech recognition technology has shown encouraging results for spoken media (audio/video) retrieval where the average precision often approaches 80% of that achieved for perfect text transcriptions. It has also been shown that the performance of the retrieval system can be severely limited by the accuracy of the recognition system depending on the quality and content oif the audio. In this paper, we present the CueVideo spoken media retrieval system where the key contributions are in developing audio prefiltering techniques to improve the accuracy of the transcript when applied to long unstructured media, and the development of an information retrieval system designed to address specific characteristics of the transcript generated using speech recognition. We also present experimental results which show that a combination of techniques we introduce can result in improved precision and recall over current spoken media retrieval techniques for real world audio.
By: S. Srinivasan, D. Petkovic, Dulce Ponceleon
Published in: RJ10143 in 1999
LIMITED DISTRIBUTION NOTICE:
This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.
Questions about this service can be mailed to reports@us.ibm.com .