User-operated speech in noise test: implementation and comparison with a traditional test

Area of Science:

• Audiology
• Speech-language pathology
• Human-computer interaction

Background:

• Assessing speech intelligibility in noise is crucial for hearing aid fitting and research.
• Traditional speech-in-noise tests can be time-consuming and require trained administrators.
• Developing user-operated tests allows for more accessible and potentially frequent assessments.

Purpose of the Study:

• To implement and evaluate a user-operated speech-in-noise test.
• To assess the efficacy of the Dantale II speech material in a self-administered format.
• To compare the performance of a user-operated test with and without a "don't know" option against a traditional test.

Main Methods:

• The study utilized the Danish Dantale II speech material, comprising five-word sentences.
• Two versions of a user-operated test were developed: one with and one without a "don't know" (?-button) option.
• Twenty-four normal-hearing subjects participated in a listening test comparing the user-operated versions to a traditional oral repetition test.

Main Results:

• Speech intelligibility, as a function of signal-to-noise ratio, was accurately described by logistic functions in all tested versions.
• Logistic parameters from the user-operated test with the ?-button closely matched those from the traditional test.
• The homogeneity of the speech material remained unaffected when used in the user-operated test format.

Conclusions:

• The Dantale II speech material is suitable for a user-operated speech-in-noise test.
• Incorporating a "don't know" (?-button) option in the user-operated test is advantageous.
• User-operated speech-in-noise tests offer a viable alternative to traditional methods.