Normative data in quiet and in noise for "DANTALE"--a Danish speech material

For a new Danish speech material, which consists of eight word lists, normative speech recognition curves are presented and the equivalence of the different lists is examined. The normative curves were obtained in quiet and in noise using three scoring methods, based on 25, 75 and 80 scoring units, respectively. Significant subject variations were found with respect to all three scoring methods due to different ways of processing the perceived sounds. Based on the test-retest SD, the two alternative scoring methods of 75 and 80 scoring units were found to be slightly more reliable than the usual word score of 25 scoring units. The equivalence of the eight word lists was accepted in quiet. Presented in noise, the difference between the lists was significant due to a different distribution of words with an initial "s" in the word lists and a significant difference in the spectra of the speech and the noise in the higher frequencies.