Article
Comparison of two procedures to measure the duration tThreshold for identifying sound samples
Search Medline for
Authors
Published: | November 28, 2019 |
---|
Outline
Text
In daily life, humans permanently perceive sounds in their surroundings that provide a multitude of different types of information, e.g. the direction, distance or velocity of a sound source helps to orientate. Another important information is the category of sound, which can be used to construct an acoustic scenery in the brain. Moreover, the identification of sounds is the basis for deciding in what way a sound needs to be interpreted, e.g. if we hear noise of a car we can try to guess the type and size of car, etc. Consequently, the identification of sounds is an important auditory process that is performed after the perception and before the interpretation of sounds. Hence, measuring the minimal duration of a sound, needed to correctly identify it, could be a useful test to analyze the auditory processing or to evaluate the hearing ability of a person.
In 2018, Sebastian Obert and Jürgen Tchorz presented a work where they used an alternative forced choice (AFC) test with four sound categories (speech, noise, animal sounds, music) to determine the duration threshold for each category. To this end, they controlled the duration of the sound samples, and measured how much the duration of the sound samples of each category can be reduced until it is not correctly identified any more. For this method, it is required that people guess, if the duration of a sound sample is too short to be identified. However, the work of Obert and Tchorz, and our own investigations indicate that people tend to have a preference, e.g. always selecting noise, if they are not sure. Therefore, in this work, another approach is analyzed to determine the duration threshold for identifying sound samples. For this alternative procedure, a sound sample is continuously repeated and the test person can adjust the duration with a slider on a touch screen. The task of the test person is to reduce the duration as long as the sound can still be correctly identified, e.g. as speech. Both procedures are tested with 20 young persons with normal hearing, and the results are compared to each other. Moreover, the advantages and disadvantages of both methods are reviewed. Furthermore, it is discussed how this alternative method can be used to analyzed sound samples, e.g. the structure or quality of speech signals.