Amplitude Compression for Preventing Rollover at Above-Conversational Speech Levels

Abstract

Hearing aids provide nonlinear amplification to improve speech audibility and loudness perception. While more audibility typically increases speech intelligibility at low levels, the same is not true for above-conversational levels, where decreases in intelligibility ("rollover") can occur. In a previous study, we found rollover in speech intelligibility measurements made in quiet for 35 out of 74 test ears with a hearing loss. Furthermore, we found rollover occurrence in quiet to be associated with poorer speech intelligibility in noise as measured with linear amplification. Here, we retested 16 participants with rollover with three amplitude-compression settings. Two were designed to prevent rollover by applying slow- or fast-acting compression with a 5:1 compression ratio around the "sweet spot," that is, the area in an individual performance-intensity function with high intelligibility and listening comfort. The third, reference setting used gains and compression ratios prescribed by the "National Acoustic Laboratories Non-Linear 1" rule. Speech intelligibility was assessed in quiet and in noise. Pairwise preference judgments were also collected. For speech levels of 70 dB SPL and above, slow-acting sweet-spot compression gave better intelligibility in quiet and noise than the reference setting. Additionally, the participants clearly preferred slow-acting sweet-spot compression over the other settings. At lower levels, the three settings gave comparable speech intelligibility, and the participants preferred the reference setting over both sweet-spot settings. Overall, these results suggest that, for listeners with rollover, slow-acting sweet-spot compression is beneficial at 70 dB SPL and above, while at lower levels clinically established gain targets are more suited.

Keywords: amplification; hearing aids; individual differences; speech perception.

Figure 1.

(a) Audiograms of the 16 test ears. The thick line shows the mean audiogram and the error bars show ±1 SD. (b) Boxplot of the PTA data for the 16 test ears and individual datapoints. (c) Boxplot of the differences between the air- and bone-conduction thresholds averaged across 500, 1000, and 2000 Hz and individual datapoints. (d) Age data of the participants. In all boxplots, the edges show the 25th and 75th percentiles and the whiskers show the range of the data. Datapoints located farther than 1.5 times the interquartile range away from the edges are shown by “+” symbols.

Figure 2.

Left: Gain curves for the NAL-NL1 setting for three ISTS input levels (55, 65, and 80 dB SPL). Middle: Same as in left panel but for the Sweet-Spot setting. Right: Gain differences between the NAL-NL1 and Sweet-Spot settings. All values show mean data across the 16 test ears together with ±1 standard error.

Figure 3.

Boxplots summarizing the DAT scores measured in quiet with the three compression settings (Sweet-Spot Fast, NAL-NL1, Sweet-Spot Slow) and two presentation levels (low, high) per setting.

Figure 4.

Scatterplots of individual raw (left) and RAU (right) scores from the DAT measurements obtained at the two presentation levels. The dashed line in each panel is the identity line. The solid lines in the left panel show the 95% CIs derived from the binomial distribution of the number of successes per 40 trials.

Figure 5.

Left: Boxplots of the HINT scores for the three compression settings. Right: Scatterplot of the differences in gain prescribed by the Sweet-Spot Slow and NAL-NL1 settings for a 70-dB-SPL input level (ΔG₇₀; horizontal axis) plotted against the differences in HINT scores obtained with these two settings (vertical axis). The solid horizontal line indicates no difference (0 dB) between the two HINT scores. Negative values correspond to less gain (horizontal axis) and lower SNRs (vertical axis) for Sweet-Spot Slow than for NAL-NL1.

Figure 6.

Left: Mean BTL scores for each of the three compression settings tested in quiet (black) or noise (blue). For each of these conditions, the BTL scores sum to 1 and can be interpreted as the probability of choosing a given setting over the other settings. Right: Preference for Sweet-Spot Slow over NAL-NL1 (in noise), as a function of the difference in gain prescribed by Sweet-Spot Slow and NAL-NL1 for a 75-dB SPL input level (ΔG₇₅). A 50% score indicates no preference.