Research | January 2020 Hearing Review

By Larry Medwetsky, PhD, Katie Kelly, AuD, and Matthew Bakke, PhD

This study revealed that for some individuals with a mild-to-moderate high frequency hearing loss, the use of an amplifier app can result in enhanced listening performance—and even performance within normal limits in background noise. However, the quality of the earphones used does make a difference. These apps and earphones may be particularly useful for these patients who—for whatever reason—could benefit from amplification but are not yet ready for a hearing aid.

In recent years, downloadable applications (apps) have become available to transform iPhone and Android mobile devices into personal sound amplifiers.1Amplifier apps typically use the microphone of a mobile device to pick up sounds and then apply processing to the incoming signal so that the individual can hear more clearly and easily.

There are a wide variety of amplifier apps; some only increase the overall volume, while others provide similar volume-control options found within a basic hearing aid. Downloadable amplifying applications are a growing phenomenon for clients with mild hearing loss for whom:

1) Hearing aids are not yet appropriate, yet would benefit from some additional amplification; 

2) Finances are a major concern and they are not ready to commit to the expense entailed by hearing aids, or 

3) For whatever reason, are simply are just not ready to try a hearing aid.

Even a patient with the mildest hearing loss can still experience consequences associated with untreated hearing loss, including a myriad of psychological (eg, reduced self-esteem, anger, depression, increased anxiety), psycho-social (eg, social withdrawal, socially inappropriate behavior or responses, lack of concentration),2,3cognitive decline issues,4 or other associated physiological issues.5 

Hearing professionals should understand how these hearing apps work so they can demonstrate them to patients who are not ideal candidates for hearing devices at a particular point in time but might later encourage them to return to this same professional when they are ready to purchase hearing devices. 

A study by Amlani et al6 compared two smartphone-based apps with that of an economy-line digital hearing aid on a number of measures, including speech-in-noise testing. Note that for this study the advanced features for the digital hearing aid were disabled to avoid biasing subjects stemming from differences in signal processing capabilities. The study examined individuals with bilateral mild-to-moderate hearing loss, with the authors finding that the smart-phone amplification apps offered some of the same performance benefits as does a “bare-bones” hearing aid. That is, there were similar electroacoustic results between the smartphone apps and the hearing aid (with advanced features turned off). In addition, speech-in-noise performance was similar across all three aided conditions. The authors concluded that results from their study indicate that smartphone-based hearing aid apps can serve as a starter solution, providing temporary assistance to individuals with hearing deficits. 

The type of receiver (ie, earphone) being used with a downloadable amplification app has not been considered relative to the application’s performance as an amplifier. Earphones come in all different shapes and sizes: from in-the-ear to on-the-ear to over-the-ear. At the time of this study, the most widely used type of earphone with smartphones was that of the in-the-ear (earbud) style. The “best” type of in-the-ear earphones to use with downloadable listening applications is unknown. If there is no difference between more expensive earphones and the earbuds a user typically receives when buying an iPhone or Android Phone, then there is no reason why an individual should invest significantly more money on the “special made” earphones. On the other hand, if the type of earphones does make a difference in performance and a client were to decide to use a downloadable amplifying application, a hearing care professional should then direct these clients toward more appropriate earphones. 

Overall, studies on smartphone software-based amplification apps are limited due to their recent arrival to the audiological world. Even less research exists on the appropriate type of in-the-ear earphone to use with these smartphone-based applications. 

This study attempted to address the latter by comparing three different brands of in-the-ear earphones, with the goal of determining whether performance varies depending on the type of in-the-ear earphone used. 

Methods

A total of 12 adults between the ages of 21 and 60 years of age (mean= 46.5; 4 females and 8 males) participated in the study. All subjects had a high-frequency (ie, >2 kHz) mild-to-moderate sensorineural hearing loss and had expressed some difficulty in everyday listening settings but had not yet received a recommendation for hearing aids. Additional inclusion criteria consisted of being a native US English speaker with a high school education. Subjects were recruited via flyers at audiology practices in the Greater Washington region, as well as from university clinics (Gallaudet University and University of Maryland). All testing was completed within one 2-hour session at the Gallaudet University Hearing and Speech Center, with subjects compensated for their time. 

Materials.The Jacoti ListenApp, which is a free, downloadable app developed by Jacoti Hearing Technologies (https://www.jacoti.com/#apps) and is reportedly the world’s first “FDA approved registered medical standalone software hearing aid,” was used as the basic amplification device on an iPhone 6. This application has been designed to assist individuals with mild-to-moderate hearing loss and is audiogram-based (ie, hearing testing must be conducted by a hearing care professional). The Jacoti ListenApp consists of left- and/or right-ear programming, four different listening settings (natural sound, speech, music, and movie), and an overall volume control, all of which can be controlled by the user. 

The three sets of in-the-ear earphones used in this study were the Apple EarPods that came with the iPhone 6, the BlueEverBlue Model 1200, and the Sennheiser IE 60. The following are the manufacturer cited specs: 

1) Apple EarPods have a frequency response range from 5 Hz to 21,000 Hz and an impedance of 45 Ohms. 

2) BlueEverBlue Model 1200 has a frequency response range from 18 Hz to 20,000 Hz and an impedance of 16 Ohms.

3) Sennheiser IE 60 earphones have a frequency response range from 5 Hz to 19,000 Hz and an impedance of 16 Ohms.

Low impedance requires less power and is better suited for small output devices such as smartphones or MP3 players. It should be noted that at the time of this study, the cost of the Sennheiser earphone was $150, while the retail cost of the BlueEverBlue model was $109.

The QuickSIN Speech-in-Noise test7,8 was used to measure subject’s performance with each set of earphones in noise. The QuickSIN is composed of a list of six sentences with five key words per sentence presented in four-talker babble noise. It requires the participant to listen to prerecorded sentences with decreasing levels of signal-to-noise ratios (SNRs, decreasing in 5 dB steps from 25 to 0 dB) and respond verbally by repeating as much of each sentence that they can. One point is given for each key word repeated correctly, with the total number of key words determining the specific SNR loss for that individual for that list. The measurement of SNR loss is important because speech understanding in noise can’t be reliably predicted from the pure-tone audiogram.9 The greater the SNR loss, the more difficulty the individual is expected to have in noisy environments. Each subject was presented two lists, and the average of these two comprised the subject’s SNR loss.

The audiobook The Hunger Games, by Suzanne Collins, was used as a subjective test measure. American actress Carolyn McCormick was recognized by AudioFile as the best voice for audiobooks in young adult literature for her work on the Hunger Games Trilogy

Procedures. An electroacoustic analysis was conducted on each of the earphones to determine their output characteristics before participant testing was initiated. Each device was monaurally fit on a KEMAR mannequin and two real-ear measurements were taken using a Bruel & Kjaer sound-level meter using a G.R.A.S. Sound and Vibration Type 40 AC microphone. Harmonic distortion of each earphone was also determined using a 1000 Hz signal. Total Harmonic Distortion (THD) indicates the degree to which the selected earphones introduce additional unwanted harmonics into the amplified signal. Spectral comparisons were also completed for each set of earphones, using white noise as an input signal with the iPhone set to maximum output. 

In assessing the subjects, otoscopic inspection and tympanometry was initially performed to ensure there were no outer or middle-ear contraindications to their participation in this study. Air conduction threshold testing was conducted in a sound-treated suite at the Gallaudet University Hearing and Speech Center using Telephonics TDH-39 supra-aural earphones connected to a diagnostic audiometer (Grason-Stadler, Model 61). Thresholds were obtained at octave and inter-octave intervals from 500 Hz through 8000 Hz to determine each participant’s threshold levels. 

The first 6 subjects were assessed only in the aided test condition (ie, using the iPhone 6 with the Jacoti amplifier app). During the course of the study, it was decided that for the last six subjects they would also be assessed in an unaided condition (ie, without the iPhone and amplifying app). Thus, a total of 12 “aided” subjects (including the six “unaided” subjects mentioned) were administered the QuickSIN listening test and three subjective listening activities in the sound field, with the participant facing one soundfield speaker, while the last 6 of the 12 subjects were also assessed in the unaided condition on the QuickSIN test while facing the soundfield speaker. Aided testing was conducted using the Jacoti ListenApp programmed on the iPhone to the subjects’ specific thresholds, the iPhone being paired with each of the three in-the-ear earphones (the specific order of earphones differed among subjects). The QuickSIN was administered at 70 dB HL per test guidelines. 

The other listening activity required the participant to listen to 1-2 minutes of The Hunger Games audiobook on a CD with a female talker at 50 dB HL and rate from 1 to 10 (1 being the worst, 10 being the best) for “sound quality” and “clarity,” and also rate from 1 to 10 the “degree of mental effort expended” (1 being great mental effort, 10 being little mental effort) of the passage using an administer-made survey. The order in which the QuickSIN and the rating activities were administered were randomized to discourage any bias due to the order of testing. Total participation time was about one hour. 

Data analysis. The data analyses consisted of administering Repeated Measures Analysis of Variance to analyze the main effects of earphone performance on the QuickSIN test. In addition, rating scales were administered to derive subjective ratings for quality, clarity, and degree of mental effort expended during the listening activity to the audiobook. These latter results were assessed via non-parametric analyses of variance.

Results

Analyses of the Total Harmonic Distortion of the three sets of in-the-ear earphones (Table 1) show that the Apple EarPods had the least amount of harmonic distortion (0.224%), followed by Sennheiser earphones (0.448%), with the BlueEverBlue earphones (0.697%) exhibiting the most amount of harmonic distortion. However, at these distortion levels it is unlikely that the listener would detect this level of distortion. Dillon and Macrae10have suggested that THD should be less than 10% and preferably less than 5%, while Killion et al11has suggested that the maximum THD should be less than 2% between 50 dB SPL and 90 dB SPL. Consequently, even though the Apple EarPods have slightly less distortion than the other earphones, this likely would not have had any impact on the test results. 

In Figure 1, with the iPhone turned to maximum output, it can be seen that the intensity of the signal across frequencies was greatest for the Sennheiser earphones, followed by the BlueEverBlue, with the differences between these two earphones and the Apple EarPods being especially pronounced below 1 kHz. Thus, the Apple EarPods exhibited significantly less bass energy than the two other earphones, appearing to display low-cut filtering below 1 kHz. Above 1 kHz, the BlueEverBlue revealed a spectrum that was similar to the Apple EarPods, though the BlueEverBlue did provide an output that was approximately 10 dB greater than the EarPods at around 16 kHz (with an output that approached that of the Sennheiser earphone at this frequency), with the Sennheiser earphones generally providing greater output than either of the other two products (ie, as high as 15 dB more than the other two earphones at a frequency region encompassing 5 kHz).

Figure 1. Spectrum comparison of a white noise signal using third octave band spectra.
Table 1. Harmonic Distortion Levels for three Earphones with iPhone set at maximum output.

Figure 2 shows the means and associated standard errors for the last six subjects on the QuickSIN test for each of the earphone “aided” conditions, as well as the No-Earphones “unaided” condition. The data reveals that these six subjects generally benefitted from the use of earphones, with subjects performing best using the Sennheiser earphones, followed by BlueEverBlue, and, in turn, the Apple EarPods. A one-way repeated measures analysis of variance was highly significant (F (3, 15) = 107.6; p= .0000). Table 2 summarizes the means, associated standard errors, as well as the 95% confidence level (lower and upper bounds) for each of these earphone conditions. The results show that performance for both the Sennheiser and BlueEverBlue earphones were significantly better than the No-Earphones condition, while mean SNR performance with the Apple EarPods—although being lower (ie, better) than that of the No-Earphones condition—did not result in significantly improved performance at the 95% confidence level. 

Figure 2. Mean performance (in dB SNR loss) for the last six subjects on the QuickSIN for each of the different earphone conditions, including when no earphones were worn.

Table 2. Mean performance on the QuickSIN for the last 6 subjects in the various earphone conditions as well as associated standard errors and 95% confidence levels.

Figure 3 displays the mean performance and associated standard errors for all 12 subjects on the QuickSIN in each of the three test conditions when earphones were worn. Results from a one-way repeated measures analysis of variance reveals that mean performance with the three earphones on the QuickSIN were significantly different from each other (F (2, 22) = 78.95; p=. 0000). Table 3a shows the means, standard errors, and levels of confidence for the three “aided” earphone conditions for the QuickSIN, while Table 3b displays the results of subsequent tests of within-subject contrasts (ie, comparison of each earphone condition against each other). The results on the QuickSIN test, as well as the subsequent subject within-contrasts tests, reveal that the participants performed significantly better with the Sennheiser and the BlueEverBlue earphones compared to the Apple earphones, and that performance on the Sennheiser earphones was significantly better than that attained on the BlueEverBlue Earphones. 

Figure 3. Mean performance for the last six subjects on the QuickSIN for each of the different
earphone conditions.

Table 3a. Mean performance on the QuickSIN test for all 12 subjects in each earphone worn condition as well as associated standard errors and 95% confidence levels.

Table 3b. Summary of QuickSIN subsequent tests of within-subject contrasts within each repeated measures analysis of variance. Note that because only two subsequent comparisons could be made within one analysis of variance, a second ANOVA was carried out to determine the within-subject contrasts for the Sennheiser versus BlueEverBlue test conditions.

Based on the various analyses, subjects generally performed better on the QuickSIN when using the iPhone with the Jacoti amplifier app as opposed to the No-Earphones condition, though there was no significant difference in performance when subjects wore the Apple EarPods versus the No-Earphones condition. Subjects performed best with the Sennheiser earphones, followed by usage of the BlueEverBlue earphones.

In addition to the speech perception measures attained on the QuickSIN test, subjects were also asked to rate the sound quality, sound clarity, and the degree of mental effort expended when listening to an audiobook for each of the earphones worn with the iPhone. Because the responses involved ordinal ratings, non-parametric analyses were conducted (specifically, the Related-Samples Friedman’s Two-Way Analysis of Variance by Ranks). Table 4 displays a summary of the findings from each of these analyses. The results show that for each rating measure, the null hypothesis was rejected and that the distributions of the ratings were not the same. Figures 4-6 display the group means and associated standard errors for sound quality, sound clarity, and mental effort respectively for each of the three earphone conditions. As was the general trend, they indicate that, for each rating measure, subjects rated performance as being the highest with the Sennheiser earphones, followed by BlueEverBlue, with performance using the Apple EarPods uniformly being the poorest on all three rating measures. 

Table 4. A summary of the findings for each of the three Rating Measures as a function of each of
the three earphone conditions (Sennheiser, BlueEverBlue, Apple EarPods) using the non-parametric Related-Samples Friedman’s Two-Way Analysis of Variance by Ranks.

Figure 4. Group mean ratings and associated errors for Sound Quality for each of the three earphone conditions. The higher the rating, the better the perceived sound quality.

Figure 5. Group mean ratings and associated errors for Sound Clarity for each of the three earphone conditions. The higher the rating, the better the perceived sound clarity.

Figure 6. Group mean ratings and associated errors for Mental Effort for each of the three earphone conditions. The higher the rating, the less the amount of mental effort perceived to have been expended.

The purpose of the present study was to determine whether the type of in-the-ear earphone used with a smartphone-based amplifier application impacts the results obtained on various measures of speech perception. Three different brands of in-the-ear earphones were compared. Output measurements with the iPhone turned to maximum output showed that the Apple EarPods exhibited significantly less low-frequency output below 1 kHz than either the Sennheiser or BlueEverBlue earphones. At frequencies above 1 kHz, the Sennheiser earphone had a frequency response that was uniformly higher than either the BlueEverBlue or Apple EarPods, with the difference being greatest in the 5 kHz region (about 15 dB greater than the other two earphones), while BlueEverBlue revealed approximately a 10 dB greater output between 15-20 kHz than the Apple EarPods, with an output level approaching the Sennheiser earphone in this frequency region.

Discussion

In comparing performance using the iPhone with the Jacoti ListenApp versus performance in the No-Earphones condition, the six subjects who were assessed revealed significantly better performance using the ListenApp with either the Sennheiser or BlueEverBlue earphones in the QuickSIN test condition, while performance with the Apple EarPods was not significantly different than that attained in the No-Earphones condition. The latter finding might be due to the limited sample size of six subjects, and it is possible that had all of the subjects been assessed in the No-Earphones condition, the power of a larger sample size might have revealed that using the Apple EarPods also resulted in significantly better findings than the No-Earphones condition. 

The frequency response of the Apple EarPods is likely the reason for the poorer outcomes on the QuickSIN and ratings on the various subjective tests, while the frequency response of the Sennheiser earphones explains why subjects performed best with these earphones. On all measures (speech recognition as well as subjective ratings), performance with the Sennheiser earphones was significantly better than either the BlueEverBlue earphones or the Apple EarPods. Subjects performed significantly better with the BlueEverBlue earphones than the Apple EarPods on the QuickSIN Test, as well as on all of the rating measures. 

These results suggest that:

1) A downloadable app, such as the Jacoti ListenApp connected to an iPhone, can result in an enhanced performance on various speech recognition and rating measures, and 

2) The type of earphones used with the Jacoti ListenApp doesmatter. The findings show that higher quality earphones than those provided with the purchase of an iPhone 6 (ie, Apple EarPods) did result in better performance and that, among the three earphones used in this study, subjects performed uniformly best with the Sennheiser, followed by the BlueEverBlue.

An even more important finding—one that was not originally an intent of this study—is the degree of enhancement that was observed with the use of the Jacoti ListenApp and Sennheiser earphones as opposed to the No-Earphones condition. The six subjects assessed on the QuickSIN in the No-Earphones condition revealed a mean SNR loss of 6.17 dB (ie, at the upper end of a mild SNR impairment if a mild SNR loss is defined as ranging from 3-7 dB per the Etymotic Research QuickSIN User Manual8). Use of the Sennheiser earphones resulted in a mean SNR loss of 1.58 dB (classified by Etymotic Research as falling within the normal range) for the six individuals who were also assessed in the No-Earphones condition (mean SNR loss of 1.25 dB across all 12 test subjects). Thus, use of the iPhone with the Jacoti ListenApp with the Sennheiser earphones resulted in an enhancement of 4.59 dB for the six subjects against the No-Earphones condition—a resultant performance in background noise comparable to that attained by listeners with hearing within normal limits. 

These results imply that similar individuals who are experiencing hearing-related difficulties but have not yet received an audiological recommendation for hearing aids or who are not yet ready to purchase hearing aids might be able to derive important listening benefits from a downloadable amplifier app (eg, Jacoti ListenApp) when used with high quality earphones. This can serve as a cost-effective recommendation for such individuals until they are recommended or are ready to purchase prescribed hearing aids. 

Conclusions and Future Directions

As mentioned, one of the limitations of this research study was the size of the sample. The non-significant results obtained when comparing Apple EarPods to the No-Earphones condition might have been due to the limited sample size of the six subjects. It is possible that, had all subjects been assessed in the No-Earphones condition, the larger sample size might have revealed that subjects wearing the Apple EarPods also performed significantly better than the No-Earphones condition. However, this still does not change the fact that the two other earphones resulted in significantly better findings than that of the Apple EarPods, with performance on the Sennheiser earphones being followed in performance by the BlueEverBlue earphones. 

Publishing such findings (perhaps on a central internet site) could enable individuals seeking to use a Smart Phone with an installed amplifier app to decide how much they are willing to spend on a set of earphones relative to the degree of speech perception enhancement achieved. In addition, audiologists and hearing aid specialists should consider ways in which they can recommend and/or dispense such devices to their non-hearing aid wearers. 

Perhaps most importantly, when subjects used the Sennheiser earphones with the iPhone, they attained hearing performance on the QuickSIN that was within normal limits. In view of the fact that the listening difficulty most often mentioned by those with mild-to-moderate hearing loss is inability to hear well in background noise, the attainment of an SNR loss that is within normal limits is an important finding. It is not known how these subjects would have performed with hearing aids (even with advanced features turned on) using the same test conditions; future research should look at such comparisons. 

One advantage that high-quality earbuds possess that current hearing aids do not is that they have an extended frequency response out to 20 kHz. The Jacoti ListenApp only provides hearing loss compensation to audio signals arriving from the microphone, without many of the processing schemes found in current digital hearing aids. Thus, the enhanced performance by using the Sennheiser earphones (and BlueEverBlue earphones to a lesser extent) likely was due to the frequency response of the earphones. The enhancement might be due to the greater degree of amplification in the 3-6 kHz region for the Sennheiser earphones as opposed to the other two earphones, or there may even be an added contribution from the Sennheiser’s enhanced frequency response at 10-16 kHz. That is, is it possible that the extended frequency response allowed the listeners to hear the high-frequency speech sounds (such as /s/, /f/, and /th/) in the presence of cocktail babble? If so, this would suggest exploring the benefits of an extended frequency response even beyond the 10 kHz that is being currently implemented in many hearing aids. 

Further research could also include the assessment of individuals using the Jacoti ListenApp (or similar amplifier apps) with various earphones as a function of different hearing loss configurations. 

Since this study, updates to the iPhone have taken place whereby wireless Bluetooth earphones, known as AirPods, are being dispensed. A number of internet sites mention the use of the AirPods with the iPhone as a hearing amplification device. Even though the AirPods come with embedded microphones, the latter are solely for use in talking on the phone and for communicating with Siri. Thus, the iPhone’s microphone remains the source for input when the iPhone + AirPods are being used as an amplification device. One possible advantage of the new AirPods, as opposed to the earphones used in this study, is that the AirPods are wireless and may address the cosmetics issue for individuals concerned with wearing corded earbuds in everyday listening situations; that is, even though the AirPods project out of one’s ears, because of their increasingly common usage individuals may not attach the same stigma to these devices as they do to hearing aids or corded earbuds. However, listening performance using the AirPods in tasks such as those administered in this study have not yet been published. One obvious question is whether the AirPods’ frequency response is similar to that of the Apple EarPods or whether they have improved. Thus, future research similar to what was conducted in this study should be carried out with the AirPods. 

Finally, if individuals are not ready to purchase hearing aids and are seeking a low-cost solution and already have a iPhone or smart phone, then downloading a free or inexpensive amplifier app and purchasing high quality earphones is probably a cheaper alternative than purchasing personal sound amplification systems (PSAPs) or future over-the-counter (OTC) hearing aids—although it remains to be seen what amplification characteristics these OTC hearing aids will possess. And, based on the results of this study, they would likely offer superior performance to that of low cost PSAPs (and possibly, the OTC hearing aids as well)—especially in noise, which is the major complaint of individuals with mild-to-moderate hearing loss.

Conclusions and Clinical Implications

There have been few studies of smartphone-based amplification applications due to their recent arrival to the audiological world. Even less research exists on the appropriate type of in-the-ear earphones to use with these smartphone-based applications. This study revealed that for individuals with a bilateral mild-to-moderate high frequency sensorineural hearing loss:

1) Downloading and using an amplifier app (eg, Jacoti ListenApp) can result in enhanced listening performance. 

2) The type of earphones used with the app does matter. 

3) The use of higher quality earphones with an effective downloadable app can result in performance that is within normal limits in background noise—at least based on the findings obtained in this study. 

Acknowledgements 

We would like to thank Bentley Plummer from BlueEverBlue for providing the BlueEverBlue earphones for this study and the impetus for assessing earphone performance when used with a smartphone. 

References

1. Medwetsky L. Mobile device apps for people with hearing loss: Part 1. Hearing Loss Magazine. 2015; September-October:20-29.

2. Ciobra A, Bianchini C, Pelucchi S, Pastore A. The impact of hearing loss on the quality of life of elderly adults. Clin Interventions Aging. 2012;7:159-163.

3. Mick P, Kawachi I, Lin FR. The association between hearing loss and social isolation in older individuals. Otolaryngol-Head Neck Surg. 2014;150(3):378–384. 

4. Lin FR, Yaffe K, Xia J, et al. Hearing loss and cognitive decline in older adults. JAMA Internal Med. 2013;173(4):293-299.

5. Abrams H. Hearing loss and associated comorbidities: What do we know? Hearing Review. 2017;24(12):32-35.

6. Amlani AM, Taylor B, Levy C, Robbins R. Utility of smartphone-based hearing aid applications as a substitute to traditional hearing aids. Hearing Review. 2013;20(12)[Dec]:16-23.

7. Killion MC, Niquette PA, Gudmundsen GI, Revit LJ, Banerjee S. Development of a quick speech-in-noise test for measuring signal-to-noise ratio loss in normal-hearing and hearing-impaired listenersJ Acoust Soc Am. 2004;116(4):2395-2405. 

8. Etymotic Research. QuickSIN Speech in Noise Test Manual, v 1.3. https://www.etymotic.com/downloads/dl/file/id/259/product/159/quicksin_user_manual.pdf.

9. Killion MC, Niquette PA. What can the pure-tone audiogram tell us about a patient’s SNR loss? Hear Jour.2000;53(3): 46-53. 

10. Dillon H, Macrae J. Derivation of design specifications for hearing aids. National Acoustics Laboratories Report Number 102. Canberra, Australia: Australian Government Publishing Service;1984. 

11. Killion MC, Van Halteren A, Stenfelt S, Warren DM. Hearing aid transducers. In: Popelka GR, Moore BCJ, Fay RR, Popper AN, eds. Hearing Aids (Springer Handbook of Auditory Research 56). 1st ed. New York, NY: Springer Publishing; 2016.

CORRESPONDENCE can be addressed to Dr Medwetsky at: larry.medwetsky@gallaudet.edu.

Citation for this article: Medwetsky L, Kelly K, Bakke M. Earphone models for iPhones: Surprising results when used with a hearing app. Hearing Review. 2020;27(1):22-25.