Objective: Of the German language speech audiometry tests used for children, only the Oldenburg children's rhyme test is constructed to allow measurements in noise. Unfortunately, no reference values exist for such measurements. Many of children admitted to test speech discrimination in noise are suspected of suffering from central auditory processing disorders and/or attention deficits. Thus, the objective of the present study is the development of a speech-in-noise test that focuses the children's attention on the target words.
Design: A speech-simulating noise was added to the Oldenburg children's rhyme test and the discrimination function was estimated. First, the total power of all target words was equalised and the total power of the noise adjusted to obtain a constant signal-to-noise ratio (SNR) for all target words. Second, speech and noise were presented binaurally with headphones. Speech was presented at 65 dB SPL and the noise level varied to obtain SNRs of 0, -5 and -10 dB SN. A total of 55 normal children, 7-10 years of age, were also tested.
Results: Mean speech discrimination was 91+/-7% at 0 dB SN, 75+/-11% at -5 dB SN and 55+/-14% at -10 dB SN. The discrimination function results in a speech reception threshold (SRT) of -10.9 dB SN and a maximum slope of 4.95%/dB SN.
Conclusion: The "Regensburg" variation of the "Oldenburg children's rhyme test" was easily feasible in all tested children. The results can be used as normative data for children between 7 and 10 years.