Abstract
The CLEF-2006 Cross-Language Speech Retrieval (CL-SR) track included two tasks: to identify topically coherent segments of English interviews in a known-boundary condition, and to identify time stamps marking the beginning of topically relevant passages in Czech interviews in an unknown-boundary condition. Five teams participated in the English evaluation, performing both monolingual and cross-language searches of speech recognition transcripts, automatically generated metadata, and manually generated metadata.
Results indicate that the 2006 English evaluation topics are more challenging than those used in 2005, but that cross-language searching continued to pose no unusual challenges when compared with monolingual searches of the same collection. Three teams participated in the monolingual Czech evaluation using a new evaluation measure based on differences between system-suggested and ground truth replay start times, with results that were broadly comparable those observed for English.