Conversational Telephone Speech Corpus Collection for the NIST Speaker Recognition Evaluation 2004

This paper discusses some of the factors that should be considered when designing a speech corpus collection to be used for text-independent speaker recognition evaluation. The factors include telephone handset type, telephone transmission type, language, and (non-telephone) microphone type. The paper describes the design of the new corpus collection being undertaken by the Linguistic Data Consortium (LDC) to support the 2004 and subsequent NIST speech recognition evaluations. Some preliminary information on the resulting 2004 evaluation test set is offered. Presented at the International Conference on Language Resources and Evaluation (4th) held in Lisbon, Portugal on 26-28 May 2004. Published in the Proceedings of the International Conference on Language Resources and Evaluation (4th), p587-590, 2004. Prepared in cooperation with University of Pennsylvania; MIT Lincoln Lab; and Federal Bureau of Investigation.