Designing a system for Swedish spoken document retrieval
Botond Pakucs, Björn Gambäck · DSpace repository (University of Tartu) · 1999
It is only during the last few years that attention has started to shift from pure textbased retrieval towards other media.Information retrieval from spoken documents is analogous to text-based retrieval; however, accessing audio documents causes some extra problems, in particular with respect to document segmentation, choice of in dexing features, and robustness.In addition, retrieval of documents in Swedish, like most non-English languages, adds the extra dimension of morphology; also, when analysing spoken Swedish data, prosodic patterns have to be taken into cu:count.In this paper we introduce SIREN, the Swedish Information Retrieval Engine, a very flexible, modular IR system which has been designed with a specific eye towards these issues.