Challenges and methods to design a child-appropriate speech-in-noise experiment in spatial auditory environments for young children

Abstract

The assessment of noise effects in controlled listening experiments for young children (aged three to six years) is a challenging topic. Tasks must be designed appropriately for this age group. Sound reproduction methods must consider children’s smaller anthropometric head dimensions to deliver plausible spatial impressions of close-to-real life scenarios. This work focuses on the objective assessment of listening effort. For this, we developed a dual-task paradigm for three to six years old children. The primary task is a double-word recognition task. The secondary task is a memory task, which consists of recalling the correct order of five symbols. This paradigm was integrated into an experimental framework including plausible sound reproduction via headphones using individual headphone transfer function equalization and individualized head-related transfer functions (HRTFs), where the interaural time difference (ITD) is adjusted according to the head dimensions of the participants. We will present preliminary results from a pretest with adults, addressing the possible effects of different noise types (speech-shaped noise, irrelevant multi-talker babble, and relevant multi-talker babble). Further, we will discuss challenges and methods within the experiment’s design process.

Publication
[Invited talk] SPIN 2022 (January 20-21, 2022), Virtual meeting