Quick SIN and BKB-SIN, two new speech-in-noise tests permitting SNR-50 estimates in 1 to 2 min

Abstract

The evolution of two new speech-in-noise tests is described, with particular emphasis on obtaining equivalency across lists. The Quick SIN Test uses IEEE sentences with female talker (L. Braida, personal communication on Harvard Phonetically Balanced Sentences developed at Harvard University during WWII, 2000); the BKB Test uses British Childrens sentences with male talker (Jon Shallop) [J. Bench and J. Bamford (eds.), Speech-Hearing Tests and the Spoken Language of Hearing-Impaired Children (Academic, London, 1979)]. In both cases, the Auditec of St. Louis four-talker babble recording was used to provide competing noise, and SNR-50 (signal-to-noise ratio for 50% correct key words in sentences) was estimated using the Tillman–Olsen recommendation [T. W. Tillman and W. O. Olsen, Modern Developments in Audiology, 2nd ed. (Academic, New York, 1973), pp. 37–74]. Normal-subject testing as well as hearing-impaired and cochlear-implant subject testing was used to check for equivalency across a range of signal-to-noise ratio performances. The standard deviation for single and multiple lists will be reported.