Self-Adjusted Amplification Parameters Produce Large Between-Subject Variability and Preserve Speech Intelligibility - PubMed (original) (raw)
Self-Adjusted Amplification Parameters Produce Large Between-Subject Variability and Preserve Speech Intelligibility
Peggy B Nelson et al. Trends Hear. 2018 Jan-Dec.
Abstract
The current study used the self-fitting algorithm to allow listeners to self-adjust hearing-aid gain or compression parameters to select gain for speech understanding in a variety of quiet and noise conditions. Thirty listeners with mild to moderate sensorineural hearing loss adjusted gain parameters in quiet and in several types of noise. Outcomes from self-adjusted gain and audiologist-fit gain indicated consistent within-subject performance but a great deal of between-subject variability. Gain selection did not strongly affect intelligibility within the range of signal-to-noise ratios tested. Implications from the findings are that individual listeners have consistent preferences for gain and may prefer gain configurations that differ greatly from National Acoustic Laboratories-based prescriptions in quiet and in noise.
Keywords: hearing aids; hearing-aid outcomes; self-fit.
Figures
Figure 1.
Long-term average spectra of the three restaurant recordings. The steady PB noise had the same long-term spectrum as the PB recording.
Figure 2.
Mean participant audiograms for left and right ears. The dashed blue lines and dotted red lines indicate 1 standard deviation from mean thresholds for left and right ears, respectively.
Figure 3.
Examples of insertion gain resulting from self-adjustment in quiet for two subjects (S1 and S18). Black dashed lines indicate insertion gain for the subject’s NAL fit. Solid orange lines are insertion gain resulting from the first self-adjustment in quiet. Dotted blue lines indicate gain resulting from a second self-adjustment in quiet. Data from S1 (upper panel) exemplify pattern of consistency in self-adjustment between repetitions of self-adjustment, while data from S18 (lower panel) reflect common pattern for subjects to reduce high-frequency gain relative to their NAL fit.
Figure 4.
Insertion gain for NAL fit and self-adjusted fits obtained in quiet. The left panel displays average gain for frequencies up to 1000 Hz, while the right panel displays average gain for frequencies between 1000 and 8000 Hz. Subjects are ordered from left-to-right on the abscissa according to the average high-frequency insertion gain in their NAL fits. Orange and blue triangles indicate, respectively, the average gain resulting from the first and second trials of self-adjustment in quiet.
Figure 5.
Test–retest histogram for insertion gain in the low- and high-frequency bands. Counts of absolute test–retest differences are shown in blue for the low-frequency band (125–1000 Hz), and in black for the high-frequency band (2000–8000 Hz).
Figure 6.
Gain deviation from NAL (self-adjusted gain minus NAL gain) for the listening conditions. Positive deviations indicate more insertion gain in the self-adjusted fit than the NAL fit. Negative deviations indicate less gain in the self-adjusted fit than in the NAL fit. Rows of plots correspond to different SNR conditions, while noise environments are indicated by marker shape and color. Gain from self-adjustment was averaged across repetitions within each condition. Subjects are ordered from left-to-right on the abscissa according to the average high-frequency insertion gain in their NAL fits (as in Figure 4).
Figure 7.
Gain deviation from NAL for individual listeners making self-adjustments in varying levels of noise. Deviations have been averaged across noise environments (within the same SNR) and repetitions. Data from low frequencies are shown in the left column; high frequencies are shown in the right. Data from individual subjects are connected with lines of varying line type; the thick, black line indicates the average deviation from NAL across subjects.
Figure 8.
IEEE key word recognition achieved using NAL (circles) and self-adjusted (triangles) fits. Large, filled symbols indicate average key word recognition across subjects. Smaller, open symbols are data from individual subjects.
Figure 9.
Difference in speech recognition performance between self-adjusted and NAL fits plotted with respect to gain deviation from NAL in the low-frequency band (bottom row) and high-frequency band (top row). Each column of panels shows data from a different SNR condition. A positive score difference indicates better performance with the self-adjusted fit than the NAL fit, while a negative score difference (lower on the ordinate) indicates worse performance with the self-adjusted fit.
Figure 10.
Histogram of IEEE key word score differences (self-adjusted fit minus NAL fit) across all subjects that completed the speech recognition assessment. Data from different SNR conditions are displayed as separate lines.
Similar articles
- Evaluation of the sparse coding shrinkage noise reduction algorithm in normal hearing and hearing impaired listeners.
Sang J, Hu H, Zheng C, Li G, Lutman ME, Bleeck S. Sang J, et al. Hear Res. 2014 Apr;310:36-47. doi: 10.1016/j.heares.2014.01.006. Epub 2014 Feb 2. Hear Res. 2014. PMID: 24495441 - Effect of slow-acting wide dynamic range compression on measures of intelligibility and ratings of speech quality in simulated-loss listeners.
Rosengard PS, Payton KL, Braida LD. Rosengard PS, et al. J Speech Lang Hear Res. 2005 Jun;48(3):702-14. doi: 10.1044/1092-4388(2005/048). J Speech Lang Hear Res. 2005. PMID: 16197282 Clinical Trial. - Exploring the limits of frequency lowering.
Souza PE, Arehart KH, Kates JM, Croghan NB, Gehani N. Souza PE, et al. J Speech Lang Hear Res. 2013 Oct;56(5):1349-63. doi: 10.1044/1092-4388(2013/12-0151). Epub 2013 Jun 19. J Speech Lang Hear Res. 2013. PMID: 23785188 Free PMC article. - An overview of the HASPI and HASQI metrics for predicting speech intelligibility and speech quality for normal hearing, hearing loss, and hearing aids.
Kates JM, Arehart KH. Kates JM, et al. Hear Res. 2022 Dec;426:108608. doi: 10.1016/j.heares.2022.108608. Epub 2022 Sep 13. Hear Res. 2022. PMID: 36137862 Free PMC article. Review. - Using statistical decision theory to predict speech intelligibility. III. Effect of audibility on speech recognition sensitivity.
Müsch H, Buus S. Müsch H, et al. J Acoust Soc Am. 2004 Oct;116(4 Pt 1):2223-33. doi: 10.1121/1.1791716. J Acoust Soc Am. 2004. PMID: 15532654 Review.
Cited by
- Preferred Strength of Noise Reduction for Normally Hearing and Hearing-Impaired Listeners.
Houben R, Reinten I, Dreschler WA, Mathijssen R, Dijkstra TMH. Houben R, et al. Trends Hear. 2023 Jan-Dec;27:23312165231211437. doi: 10.1177/23312165231211437. Trends Hear. 2023. PMID: 37990543 Free PMC article. - Efficacy and Effectiveness of Evidence-Based Non-Self-Fitting Presets Compared to Prescription Hearing Aid Fittings and a Personal Sound Amplification Product.
Venkitakrishnan S, Urbanski D, Wu YH. Venkitakrishnan S, et al. Am J Audiol. 2023 Nov 13;33(1):1-24. doi: 10.1044/2023_AJA-23-00121. Online ahead of print. Am J Audiol. 2023. PMID: 37956699 Free PMC article. - Personalizing over-the-counter hearing aids using pairwise comparisons.
Vyas D, Brummet R, Anwar Y, Jensen J, Jorgensen E, Wu YH, Chipara O. Vyas D, et al. Smart Health (Amst). 2022 Mar;23:100231. doi: 10.1016/j.smhl.2021.100231. Epub 2021 Nov 25. Smart Health (Amst). 2022. PMID: 37397910 Free PMC article. - Machine Learning-Based Hearing Aid Fitting Personalization Using Clinical Fitting Data.
Mondol SIMMR, Kim HJ, Kim KS, Lee S. Mondol SIMMR, et al. J Healthc Eng. 2022 Oct 15;2022:1667672. doi: 10.1155/2022/1667672. eCollection 2022. J Healthc Eng. 2022. PMID: 36285186 Free PMC article. - Personalization of Hearing Aid Fitting Based on Adaptive Dynamic Range Optimization.
Ni A, Akbarzadeh S, Lobarinas E, Kehtarnavaz N. Ni A, et al. Sensors (Basel). 2022 Aug 12;22(16):6033. doi: 10.3390/s22166033. Sensors (Basel). 2022. PMID: 36015791 Free PMC article.
References
- Allen, J. B., & Berkley, D. A. (1979). Image method for efficiently simulating small-room acoustics. Journal of the Acoustical Society of America, 65, 943. doi:10.1121/1.382599.
- Byrne D. (1986) Effects of frequency response characteristics on speech discrimination and perceived intelligibility and pleasantness of speech for hearing-impaired listeners. The Journal of the Acoustical Society of America 80: 494–504. doi:10.1121/1.394045. - PubMed
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Other Literature Sources
Medical