Perceptual Effects of Adjusting Hearing-Aid Gain by Means of a Machine-Learning Approach Based on Individual User Preference.

This study investigated a method to adjust hearing-aid gain by use of a machine-learning algorithm that estimates the optimal setting of gain parameters based on user preference indicated in an iterative paired-comparison procedure. Twenty hearing-impaired participants completed this procedure for 12 different sound scenarios. During the adjustment procedure, their task was to indicate a preference based on one of three sound attributes: Basic Audio Quality, Listening Comfort, or Speech Clarity. In a double-blind comparison of recordings of the processed scenarios, and using the same attributes as criteria, the adjusted gain settings were subsequently compared with two prescribed settings of the same hearing aid (with and without activation of an automatic sound-classification system). The results showed that the adjustment method provided a general improvement of Basic Audio Quality, an improvement of Listening Comfort in a traffic-noise scenario but not in three scenarios with speech babble, and no significant improvement of Speech Clarity. A large variation in gain adjustments was observed across participants, both among those who did benefit and among those who did not benefit from the adjustment. There was no clear connection between the gain adjustments and the perceived benefit, which indicates that the preferred gain settings for a given sound scenario and a given listening intention are highly individual and difficult to predict.

Evaluation of the Efficacy of a Dual Variable Speed Compressor over a Single Fixed Speed Compressor.

BACKGROUND: It has been suggested that hearing-impaired listeners with a good working memory (WM) should be fitted with a compression system using short time constants (i.e., fast-acting compression [FAC]), whereas those with a poorer WM should be fitted with a longer time constant (i.e., slow-acting compression [SAC]). However, commercial hearing aids (HAs) seldom use a fixed speed of compression. PURPOSE: The performance of a variable speed compression (VSC) system relative to a fixed speed compressor (FAC and SAC) on measures of speech intelligibility, recall, and subjective report of listening effort and tolerable time was evaluated. The potential interaction with the listeners' WM capacity (WMC) was also examined. RESEARCH DESIGN: A double-blinded, repeated measures design. STUDY SAMPLE: Seventeen HA wearers (16 with greater than one year HA experience) with a bilaterally symmetrical, mild to moderately severe sensorineural hearing loss participated in the study. DATA COLLECTION AND ANALYSIS: Participants wore the study HAs at three compression speeds (FAC, SAC, and VSC). Each listener was evaluated on the Office of Research in Clinical Amplification-nonsense syllable test (NST) at 50 dB SPL (signal-to-noise ratio [SNR] = +15 dB), 65 dB SPL (SNR = +5 dB), 80 dB SPL (SNR = 0 dB), and a split (80 dB SPL-50 dB SPL) condition. Listeners were also evaluated on a Repeat Recall Test (RRT), where they had to repeat six short sentences (both high- and low-context sentences) after each was presented. Listeners recalled target words in all six sentences after they were presented. They also rated their listening effort and the amount of time they would tolerate listening under the specific condition. RRT sentences were presented at 75 dB SPL in quiet, as well as SNR = 0, 5, 10, and 15 dB. A Reading Span Test (RST) was also administered to assess listeners' WMC. Analysis of variance using RST scores as a covariate was used to examine differences in listener performance among compressor speeds. RESULTS: Listener performance on the NST was similar among all three compression speeds at 50, 65, and 80 dB SPL. Performance with FAC was significantly better than SAC for the split condition; however, performance did not differ between FAC and VSC or between SAC and VSC. Performance on the NST was not affected by listeners' RST scores. On the RRT, there was no effect of compressor speed on measures of repeat, recall, listening effort, and tolerable time. However, VSC resulted in significantly lower (better) speech reception threshold at the 85% correct recognition criterion (SRT85) than FAC and SAC. Listener RST scores significantly affected recall performance on the RRT but did not affect SRT85, repeat, listening effort, or tolerable time. CONCLUSION: These results suggest that the VSC, FAC, and SAC yield similar performance in most but not all test conditions. FAC outperforms SAC, where the stimulus levels change abruptly (i.e., split condition). The VSC yields a lower SRT85 than a fixed compression speed at a moderately high level with a favorable SNR. There is no interaction between compression speed and the participants' WMC.