The Type of Noise Influences Quality Ratings for Noisy Speech in Hearing Aid Users

Abstract

Purpose The overall goal of the current study was to determine whether noise type plays a role in perceptual quality ratings. We compared quality ratings using various noise types and signal-to-noise ratio (SNR) ranges using hearing aid simulations to consider the effects of hearing aid processing features. Method Ten older adults with bilateral mild to moderately severe sensorineural hearing loss rated the sound quality of sentences processed through a hearing aid simulation and presented in the presence of five different noise types (six-talker babble, three-talker conversation, street traffic, kitchen, and fast-food restaurant) at four SNRs (3, 8, 12, and 20 dB). Results Everyday noise types differentially affected sound quality ratings even when presented at the same SNR: Kitchen and three-talker noises were rated significantly higher than restaurant, traffic, and multitalker babble, which were not different from each other. The effects of noise type were most pronounced at poorer SNRs. Conclusions The findings of this study showed that noise types differentially affected sound quality ratings. The differences we observed were consistent with the acoustic characteristics of the noise types. Noise types having lower envelope fluctuations yielded lower quality ratings than noise types characterized by sporadic high-intensity events at the same SNR.

Figure 1.

Audiograms (averaged across both ears for each participant) are shown in gray. The black line shows the average audiogram across all participants.

Figure 2.

Provides an overview of the experimental design including the characteristics of the stimuli that entered the hearing aid microphones (noise type, SNR, and level of speech) in the stimuli recording, hearing aid simulation parameters for the mild and moderate hearing aid processing conditions, acoustic parameters such as vent and modifications of the signal by way of the ear canal, and the listener's perceptual task following headphone presentation. WDRC = wide range dynamic compression; HINT = Hearing in Noise Test; SNR = signal-to-noise ratio; HA = Hearing Aid; HP = High-pass Filter; LP = Low-pass Filter.

Figure 3.

The noisy stimuli were generated in the signal-to-noise ratios of 3, 8, 12, and 20 dB. In each stimulus, the compound sentence was A-weighted and held at 65 dBA, and the appropriate level of an A-weighted noise to achieve a given signal-to-noise ratio was calculated. Then, the two parts were combined.

Figure 4.

Scatter plot showing the distributions of the quality ratings at various SNRs (shown on the x-axis) and noise types (shown with different colors; averaged across vent, processing, participant, and segment). Each noise type is fit with a linear regression line. SNR = signal-to-noise ratio.

Figure 5.

Temporal characteristics for each noise type in isolation. The plots show a concatenation of the nine 4.1-s segments selected for each noise type.

Figure 6.

Long-term auditory spectra for the five noise types computed for nine concatenated noise segments. Each spectrum is calculated as the output of a 128-band gammatone auditory filterbank, with band center frequencies shown along the x-axis. The spectral level in each band is indicated in dB re: broadband signal level. The noise types are indicated by color. RMS = root-mean-square.

Figure 7.

Line graphs showing average listener quality ratings (y-axis) as a function of cepstral correlation (x-axis) for each of the five noise types (indicated by panel and color) at each signal-to-noise ratio (shape). The range of cepstral correlation values is indicated for each noise type with curly braces. A smaller value on the y-axis indicates lower quality rating, and a smaller value on the x-axis indicates lower cepstral correlation, that is, more difference between the test signal and reference signal.

Figure 8.

Scatter plots showing average listener quality ratings (y-axis) as a function of cepstral correlation (x-axis; left panel) and as a function of SNR (x-axis; right panel). Ratings were averaged across listeners, segments, vent, processing, and noise type. A regression line was fit for each comparison, and R2 values were calculated based on these regressions. Rating increases as cepstral correlation increases, and rating increase as SNR increases, though a higher proportion of variance in rating is explained by cepstral correlation than the proportion of variance explained by SNR alone (as indicated by the higher R2 value for the cepstral correlation comparison). SNR = signal-to-noise ratio.

Figure C1.

Cepstral correlation as a function of dB SNR. Noise type is designated by color. As signal-to-noise ratio increases, the difference in the cepstral correlation value between noise types diminishes.