Approximately 48 million people in the United States have hearing loss or hearing difficulties in noisy environments. Whisper.ai Inc has previously brought to market a commercial hearing aid system that reduces background noise and amplifies voices and sounds of interest using its proprietary platform based on machine learning and artificial intelligence technologies. Whisper.ai Inc now seeks to bring to market a new hearing system that will incorporate a "self-fitting" capability. Self-fitting hearing aids have emerged in recent years as a potentially viable option to calibrate hearing aids to the needs of individual users without clinician intervention. The purpose of this project is to evaluate the efficiency and reliability of the fitting procedure and algorithms developed by Whisper.ai Inc. The study will be carried out using a randomized crossover design in two phases: in phase 1, subjects will be tested and a hearing aid will be fitted using conventional audiological standard procedures , and in phase 2, the subject will be tested and a hearing aid will be fitted using the results of the self-fitting algorithm. Objective, subjective, and behavioral responses will be gathered from a variety of hearing-related surveys and tests, and will be analyzed quantitatively to evaluate the efficiency and reliability of the self-fitting algorithm. The investigators expect the fitting results of the Whisper.ai self-fitting algorithm to be similar to those of standardized fitting procedures conducted by hearing professionals.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
SINGLE
Enrollment
80
Whisper.ai
San Francisco, California, United States
RECRUITINGHearing aid gain comparisons using Real Ear Measures Data Analysis Plan
To evaluate, fine-tune, and compare the gains produced by the Whisper Hearing System using both PRO-FIT and SELF-FIT, general standard-of-care real-ear measurements (REM) will be used. Real Ear Measures (REMs) provide an objective measure of sound pressure level in an individual's ear canal. Frequencies critical to speech understanding (500 through 4000 Hz) will be averaged to yield a single value similar to a pure tone average. A mean absolute difference will be used to evaluate whether the SELF-FIT is non-inferior to the PRO-FIT; a mean absolute difference less than 5 dB will be considered as a non-significant difference when comparing REMs between the SELF-FIT and PRO-FIT groups.
Time frame: 4 weeks
Speech perception scores comparison
The Quick Speech In Noise (QuickSIN) will provide a behavioral measure that will be compared across fitting conditions.The outcome of a non-inferiority trial may be assessed by a two-sided t-test, using a 95% confidence interval. A two-sided t-test will be used to compare the two final Quick Speech in Noise scores across fitting groups.Minimum value: -4.5 SNR Loss Maximum value: 25.5 SNR Loss Lower score here is better performance/outcome (-4.5 is the best possible score).
Time frame: 2 hours for the testing, 4 weeks trial
Overall perception hearing
Abbreviated Profile of Hearing Aid Benefit (APHAB) is a 24-item self-assessment inventory in which subjects report the amount of trouble they are having with communication or noises in various everyday situations. The APHAB produces scores for 4 subscales: Ease of Communication, Reverberation, Background Noise, and Aversiveness. These are combined to create a global score, which will be used to evaluate the overall benefit of amplification for each fitting strategy (SELF-FIT and PRO-FIT). The outcome of a non-inferiority trial may be assessed by a two-sided t-test, using a 95% confidence interval. A two-sided t-test with 95% confidence intervals will be used to determine whether APHAB outcomes differ between the SELF-FIT and the PRO-FIT. Minimum value: 0% of problems Maximum value: 100% of problems Lower score here is a lower percent of problems, or better performance/outcome.
Time frame: 4 weeks
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.