Hearing Aid Fitting | December 2019 Hearing Review

The acoustic effects of various ear tips, and how they can influence RIC hearing aid fittings.

By Laura Winther Balling, PhD; Niels Søgaard Jensen, MSc; Sueli Caporali, PhD; Jens Cubick, PhD, and Wendy Switalski, MBA, AuD

A variety of ear tips are now available for receiver-in-the-canal (RIC) hearing aids. But how might they influence a hearing aid fitting? Analyzing real-ear measurements and vent effects, this study looks at five different kinds of RIC ear tips and their effects on the sound reaching the eardrum and resulting sound quality. The bottom line: Dispensing professionals need to understand the effects of the specific ear tip in the actual ear and recognize that using an “instant-fit tip” cannot simply be an “instant fitting,” but instead requires individualization with the same care that is given to prescribing gain.

Receiver-in-canal (RIC) hearing aids (HAs) with instant-fit ear tips have become increasingly popular over the last decade. For example, around 82% of HAs dispensed in the United States in the first half of 2019 were RIC devices.1 These are generally fit with instant ear tips, which have several advantages for both the hearing care professional (HCP) and the end user, primarily time efficiency and physical comfort. However, this type of fitting also raises some challenges, which need to be dealt with in order to ensure appropriate gain and output, and optimal sound quality for the end user.

In this article, we address the challenges that arise from variations in the direct sound and vent effect for instant-fit ear tips. When the effective vent size—including the vents in the ear tip and the leakage around it—is large, only a small amount of low-frequency sound from the hearing aid (HA) reaches the eardrum; this is what is referred to as the vent effect (VE).2 At the same time, the insertion loss (IL)—the attenuation of the direct sound reaching the ear drum caused by the presence of the hearing aid in the ear—is smaller for larger effective vent sizes.

This makes it important to understand the effective vent size of instant ear tips—something which, with only a few exceptions (eg, Mueller and Ricketts3), has not been extensively addressed in the scientific literature. In the present article, we describe and discuss a recent study4 that used extensive real-ear measurements to show how IL and VE vary substantially between users, and we outline how this variation may be accommodated in the fitting. The measurements we report are for five different Widex ear tips, but the results may be generalized to ear tips provided by other manufacturers.

A further aspect to consider—particularly for those ear tips that are open by design, but in practice for all types—is the mixing of the direct sound through the vent and/or any leakage around the ear tip with the amplified sound from the HA, which is delayed by the digital signal processing. When the two sound contributions are of approximately equal amplitude, differences in phase will lead to cancellations, which can be seen as ripples in the spectra of the resulting sound. This artifact, called the comb-filter effect, has negative effects on sound quality.

Insertion Loss and Vent Effects

Subjects. A total of 30 normal-hearing individuals (10 female, 20 male) participated in the experiment. The mean age was 45 years (range: 19 to 67 years). Measurements from one participant were excluded due to issues with cerumen occluding the probe tube, so measurements are reported for 58 ears.

Methods. The experiment consisted of a series of real-ear measurements with five different ear tips. Participants were fitted with a pair of Widex EVOKE 440 Passion RIC HAs with S-receivers and the following instant ear tips: Widex Open, Tulip, Round (2-vent), Round (1-vent), and Double domes (Figure 1). Measurements were made for all ear tips for all participants, but in different orders. During the measurements, the microphone in the HA was not used. Instead, the only active processing task of the HA was to present a sound signal streamed to the HA via the receiver into the ear canal.

Figure 1. The five different instant ear tips used in this study: a) Open; b) Tulip; c) Round 2-vent; d) Round, 1-vent; e) Double domes.

The real-ear measurements were conducted using an Interacoustics Affinity 2.0 system. The participants were seated 1 meter from the system’s loudspeaker in a standard audiometric test suite. To measure the real-ear unaided response (REUR), probe tubes were placed in the ear canal within 5 mm of the eardrum, as verified by otoscopy. Pink noise was then presented from the Affinity system at 0° azimuth at a level of 65 dB SPL. Next, without changing the position of the probe tube, the hearing aids were placed on the ears and real-ear occluded responses (REOR) were measured with the HAs switched off using the same pink noise signal. The IL was calculated as the difference between the REOR and the REUR. The IL measurement setup is shown in Figure 2, together with an example of individual results.

Figure 2. Real-ear measurement configurations used for insertion loss (IL) estimation, with pink noise presented from a loudspeaker in front of the participant. IL was calculated as the difference between REOR and REUR, as shown in the example on the right.

After conducting the measurements for the IL, and without removing the hearing aids and probe tube, another set of real-ear measurements were conducted to estimate the VE based on streaming of brown noise to the HAs via a TV-DEX accessory. The first measurement was conducted with the hearing aid streaming in the ear. Before the second measurement, and with the hearing aids still in place, the ear canal and concha were filled with impression material in order to measure the response from the hearing aids with a fully occluded ear, thereby making sure that no sound could “escape” from the ear canal. The difference between these two measurements shows the VE—defined as how much the HA sound is reduced at the eardrum when the HAs are fitted with the given ear tip. The VE measurement setup is shown in Figure 3, together with an example of individual results.

Figure 3. Real-ear measurement configurations used for VE estimation, with brown noise being streamed to the HA and presented via the receiver. VE was calculated as the difference between the ‘normal’ response and the ‘occluded’ response, as shown in the example on the right.

Results

Figure 4. Average IL in 1/3-octave bands across 58 ears for the five ear tips (top left). The other panels show the average IL per tip (thick colored line) ±1 standard deviation (color-shaded area). The grey-shaded area represents the observed range of individual measurements.

IL comparisons. Figure 4 above shows the average IL in 1/3-octave bands for each of the different ear tips in the top left panel (“All tips”), and the variation across ears for each of the five ear tips in the remaining panels. The IL can be thought of as the attenuation that the ear tip imposes on sounds from the environment, taking into account its effect on the ear-canal resonance. Two things are important in the results shown here:

  1. We see in the average IL that the ear tips fall into three distinct groups, as expected:
  • The Open ear tips are mostly transparent (ie, 0 dB IL) for sound generated outside the ear canal, apart from a slight attenuation of approximately 2 dB at frequencies above 2 kHz.
  • The Tulip and the two Round ear tips form a second group, with a transparent response up to 1 kHz and a maximum attenuation of approximately 9 dB (Tulip, Round 2-vent) to 12 dB (Round, 1-vent) at a frequency of about 2.6-2.8 kHz;
  • The Double domes are on average only transparent up to 600 Hz, and they show the highest average attenuation of 16 dB at 3 kHz.

2) We see substantial variability in IL among individual ears. The variability, illustrated by the color-shaded areas (standard deviation) and grey-shaded areas (range from minimum to maximum), is largest for the Double domes, but also considerable for the other types of ear tips.

VE comparisons. Figure 5 shows the VE data plotted in the same way as the IL data in Figure 4. Again, we see the same pattern of three groups of ear tips, and substantial variation between ears for each ear tip:

1) On average, the largest VE was found for Open tips. Here, the largest average vent loss is about 40 dB around 125 Hz, and a vent loss is observed up to just below 2 kHz.

2) The measurements for Tulip, Round (2-vent) and Round (1-vent) are again quite similar, with a vent loss going up to between 1 and 1.5 kHz, and a maximum attenuation of about 30 dB at the lowest frequencies.

3) Double domes show the least pronounced vent loss, with a cut-off frequency of about 1.2 kHz and an average VE of 24 dB around 125 Hz.

Figure 5. Average VE in 1/3 octave bands across 58 ears for the five ear tips (top left). The other panels show the average VE per tip (thick colored line) ±1 standard deviation (color-shaded area). The grey-shaded area represents the observed range of individual measurements.

However, while these numbers represent averages, measurements again show high individual variability in the VE across ear tips. Open shows the lowest variability, and Double domes show the greatest variability, as was also the case for the IL. At the lowest frequencies, the VE with Double domes can be as low as 6 dB (nearly fully closed) or as high as 38 dB (nearly completely open).

Implications for Fitting Hearing Aids

Both the IL and VE measurements showed differences between ear tips, and large inter-participant variability. Both these overall results have important implications for fitting with instant-fit ear tips.

First, it is clear (and not surprising) that the choice of ear tip has big implications for HA acoustics, both in terms of how much direct sound is transmitted (IL) and in terms of how much low-frequency sound escapes through the ear tip (VE). This means that to achieve the desired gain and output targets, and to provide optimal sound quality for the individual user, the fitting software should take the acoustics of the actual ear-tip style into account during fitting, as is done in the Widex Compass GPS fitting software.

Second, the high variability in the measurements between ears shows how crucial it is to take the individual effective vent size into account. This applies for all ear tips where both the average IL and VE may vary significantly from the actual IL and VE for the individual HA fitting (as shown in Figures 4 and 5). For example, the highest average VE for the Tulip ear tips is just over 30 dB, but in the individual measurements, the largest VE varies from below 20 dB to above 40 dB. This highlights the importance of evaluating the effective vent size for each individual fitting instead of relying on averages. In Compass GPS, this is done through the feedback test, where the effective vent size is estimated and subsequently factored into the prescription of gain by compensating for the individual’s VE. This ensures that an individual with a larger-than-average effective vent does not get too little gain in the low frequencies and a tinny sound quality, while an individual with a smaller-than-average effective vent does not get too much gain and a boomy effect.

In addition to these more general consequences, the observed VE—even for the supposedly closed Double domes—may in many cases be so large that listeners with relatively large low-frequency hearing loss do not get the prescribed gain in the low frequencies. This points to the advantage of using custom molds for such hearing losses. This unintended venting will also have negative consequences for low-frequency directional and noise-reduction processing.

Implications for Sound Quality

A further crucial challenge for sound quality in open fits is the comb-filter effect. As described above, this effect arises when the direct sound that comes through any venting and leakage mixes with the signal from the hearing aid, which is typically delayed between 2-8 ms due to the signal processing. Delays in the lower end of this range, as is seen in the Widex EVOKE hearing aids, generally give better sound quality, especially for own voice.5 However, even a delay of a few milliseconds means that the direct and HA sounds may be out of phase at certain frequencies, resulting in peaks and troughs in the signal that resemble the teeth of a comb when plotted as a gain-frequency curve. These distortions are most pronounced when the two sound sources are of approximately equal amplitude, which, depending on the amplification and the individual effects of the HA ear tips, will typically be in the range from 500 Hz to 2 kHz. Research on delays in hearing aids has focused primarily on determining the upper delay limit within which HA sound quality is tolerable,6-8 but even within this limit the artifact remains.

The mixing of sounds will be most apparent with open fits, but as the results reported above show, all the different ear tips may be effectively open for the individual ear, so the risk of comb-filtering compromising sound quality is quite substantial. A precise estimation of the vent size will allow gain to be turned down in those frequency bands that give the most pronounced comb-filter effects, while leaving non-affected frequency bands unaltered; however, this does of course need to be balanced with the aim of providing the best possible audibility for the individual user.

Conclusion

In this article, we have demonstrated some of the challenges of fitting with instant-fit ear tips, with a focus on the variation in direct sound reaching the eardrum and in vent effects, both between ear tips and between individual users. It is crucial that these acoustic variations are considered during fitting, both generally for the different types of ear tips and specifically for the individual HA user’s effective vent size. This is important in order to be able to provide the prescribed gain, but it also has great implications for sound quality—both for the tonal balance (perceptions of tinny or boomy sound quality) and for preventing the distortions that are due to the comb-filter effect. This requires understanding the effects of the specific ear tip in the actual ear and recognizing that using an “instant-fit tip” cannot simply be an “instant fitting,” but instead requires individualization with the same care that is given to prescribing gain.

References

  1. Strom KE. Hearing aid sales increase by 3.8% in first half of 2019. July 17, 2019. Available at: https://hearingreview.com/2019/07/hearing-aid-sales-increase-3-8-first-half-2019/
  2. Kuk F, Nordahn M. Where an accurate fitting begins: Assessment of in-situ acoustics (AISA). Hearing Review. 2006;13(7):34-42.
  3. Mueller HG, Ricketts TA. Open-canal fittings: Ten take-home tips. Hear Jour. 2006;59(11):24-39.
  4. Caporali S, Cubick J, Catic J, Damsgaard A, Schmidt E. The vent effect in instant ear tips and its impact on the fitting of modern hearing aids. Poster presented at: International Symposium on Auditory and Audiological Research (ISAAR), Nyborg, Denmark, August 2019.
  5. Groth J, Søndergaard MB. Disturbance caused by varying propagation delay in non-occluding hearing aid fittings. Int J Audiol. 2004;43(10):594-599.
  6. Stone MA, Moore BCJ. Tolerable hearing aid delays. II. Estimation of limits imposed during speech production. Ear Hear. 2002;23(4):325-338.
  7. Stone MA, Moore BCJ. Tolerable hearing aid delays. IV. Effects on subjective disturbance during speech production by hearing-impaired subjects. Ear Hear. 2005; 6(2):225-235.
  8. Goehring T, Chapman JL, Bleeck S, Monaghan JM. Tolerable delay for speech production and perception: Effects of hearing ability and experience with hearing aids. Int J Audiol. 2017;57(1):61-68.

ABOUT THE AUTHORS: Laura Winther Balling, PhD, is an Evidence and Research Specialist, Niels Søgaard Jensen, MSc, is a Senior Evidence and Research Specialist, and Sueli Caporali, PhD, and Jens Cubick, PhD, are Audiological Performance Specialists at Widex. Wendy Switalski, MBA, AuD, is Director of Professional Development at Widex USA.          

CORRESPONDENCE
can be addressed to Dr Balling at: [email protected]

CITATION FOR THIS ARTICLE: Balling LW, Jensen NS, Caporali S, Cubick J, Switalski W. Challenges of instant-fit ear tips: What happens at the eardrum? Hearing Review. 2019;26(12)[Dec]:12-15.