Development and Evaluation of the Russian Digit Triplet Test

The purpose of this study was to develop, optimize, and evaluate the Russian digit triplet test for hearing screening in noise. The speech material consists of sequences of three digits like 3-2-8. A closed-set test format enables its automated administration. The speech material was optimized and evaluated for headphone and landline telephone presentation. Results were compared to digit triplet tests in other languages. First, the speech material was recorded and optimized with respect to intelligibility by applying level corrections on the basis of digit-specific speech reception thresholds (SRT) defined as signal-to-noise ratio yielding 50% intelligibility. In the following, evaluation measurements were conducted to verify test list equivalency and to obtain reference data for listeners with normal hearing. Evaluation measurements showed small differences across the six test lists. Mean slopes of the list-specific recognition functions were 14.8 %/dB and 14.4 %/dB for the headphone and the landline telephone version, respectively. The reference SRT with corresponding standard deviation for the headphone and the landline telephone version were – 11.7 ± 0.9 dB and – 8.5 ± 0.6 dB, respectively. The difference of 3.2 dB in SRT between the two test versions could be explained by the limited frequency bandwidth for the transmission over landline telephone. Both versions were shown to be reliable tests for screening speech recognition in noise and to be comparable with other languages.