No abstract

People hear differently and have different aesthetic criteria. Due to a number of these reasons, individuals may prefer different processing parameters for both speech and music. For example, some individuals may prefer more low frequency emphasis and longer reverberation time while listening to music. Alternatively, certain individuals with hearing loss may prefer higher compression ratios at high frequencies. To perform this “personalization,” a number of psychometric procedures can be used. They vary in terms of accuracy, complexity, testing time, etc. For consumer devices, such as headsets, personal amplifiers, personal music players, etc., we have developed a new methodology, called EarPrint™, which is fast, intuitive, and requires minimum instructions. The procedure is based on reducing a multidimensional parametric space into a 2-D surface representation where the user perceives a consistent quality increment as he/she moves his/her finger along a straight line. In this paper, we evaluate how valid and reliable the EarPrint™ results are when compared to more rigorous psychometric procedures (e.g., category scaling or magnitude estimates.) Fifteen subjects were tested and the results were analyzed in terms of validity, dependence on listening material and reliability.

24 Frames (2017)USADirector Abbas Kiarostami Runtime 114 minutesBlu-rayUSA, 2019 Distributed by The Criterion Collection (Region A/1)Bowling for Columbine (2002)USADirector Michael Moore Runtime 120 minutesBlu-ray USA, 2018Distributed by The Criterion Collection (Region A/1)Heart of a Dog (2015)USADirector Laurie Anderson Runtime 75 minutesDVD USA, 2016Distributed by The Criterion Collection (region 1)Monterey Pop (1968)USADirector D. A. Pennebaker Runtime 79 minutesBlu-ray USA, 2017Distributed by The Criterion Collection (region 1)

In theory, a directional hearing device could apply large attenuation to a single noise source and achieve great enhancement of signal-to-noise ratio relative to an omni-directional device. However, as observed in many clinic studies, the actual benefit of a directional device measured in typical rooms rarely exceeds 6 dB, even in conditions with a single noise source located in the direction of maximum attenuation of the device. This dramatic reduction from the theoretical benefit is a manifestation of many factors, including mismatch in the microphone responses, misalignment of the device relative to the target source, head movement, head shadow, room reverberation, and so on. In the present study, we examine the relative contributions of these factors using computer simulations and electroacoustical measurements. The results provide a basis for setting realistic expectations of the benefit of directional devices for specific listening environments.