Paper: Investigating Cross-Language Speech Retrieval For A Spontaneous Conversational Speech Collection

ACL ID N06-2016
Title Investigating Cross-Language Speech Retrieval For A Spontaneous Conversational Speech Collection
Venue Human Language Technologies
Session Short Paper
Year 2006
Authors

Cross-language retrieval of spontaneous speech combines the challenges of working with noisy automated transcription and lan- guage translation. The CLEF 2005 Cross- Language Speech Retrieval (CL-SR) task provides a standard test collection to inves- tigate these challenges. We show that we can improve retrieval performance: by care- ful selection of the term weighting scheme; by decomposing automated transcripts into phonetic substrings to help ameliorate tran- scription errors; and by combining auto- matic transcriptions with manually-assigned metadata. We further show that topic trans- lation with online machine translation re- sources yields effective CL-SR.