The Acoustic Transparency of Waterproof Smartphone Cases

Abstract

Phone use is a critical communication event in many people's lives. Audiologists have aimed to assist individuals with hearing loss and phone usage through the use of technology and counseling. To counsel effectively, all contributions to hearing difficulty on the phone must be considered, including the effects of smartphone cases. The purpose of this study was to evaluate the effects on dB output caused by waterproof smartphone cases that cover the ear-level speaker. One waterproof case was tested with three smartphones, two iPhones, and one Android. A second waterproof case was tested with the two iPhones. Results revealed there was significant attenuation of the audio-signal by both waterproof smartphone cases that was great enough in one case/phone combination to potentially result in a complete lack of intelligibility of the signal.

Background/Rationale

“Mobile Technology and Home Broadband 2019” reported the prevalence and various uses of smartphones. ¹ They found that 81% of Americans own a smartphone, up from 35% in 2011. It is estimated that the use of smartphones will continue to increase, as already 37% of adults report they are using it as their primary internet access in the home. As connectivity options with personal amplification devices on the rise, persons with hearing loss are finding increased benefits from smartphone apps including streaming music, hearing alarms, transcribing speech, controlling amplification features, and advanced signal processing in noise. ² Research with apps specifically designed to enhance speech recognition in noise via the microphone of the smartphone has shown that listeners with normal and impaired hearing can identify sentences in background noise significantly better than when listening with hearing aids alone. ³

Due to the increased utility of smartphone technology, there is a competitive market for smartphone accessories, such as protective cases. Some protective cases have a waterproof feature. Most of these encase the smartphone completely, including the ear-level speaker, and have a front-and-back shell that snap together. The frame of the case is a hard plastic, with some rubber material embedded within that to facilitate the grip of the user. The top of the front piece is a thinner, transparent plastic material that covers the whole front face of the smartphone, including the screen and ear-level speaker.

Many features of smartphone protective cases have been investigated to support their marketing claims of total protection. Some of these features include their resilience to scratches, drops, and depth and duration of water submersion. One aspect not yet addressed in marketing is how the materials of the protective case, which cover the ear-level speaker, affect the spectral output of the acoustic signal for those who use the phone in the traditional way without connectivity to personal hearing assistive devices.

With approximately 15% of American adults reporting some trouble hearing, ⁴ it is important to consider all factors that could potentially create a more difficult listening situation. When intensity is reduced, speech recognition may be compromised. In 1971, Hood and Poole obtained performance intensity (PI) functions using both sentences and words for individuals with normal and sensorineural and conductive hearing loss. ⁵ They found that percent-correct recognition scores increased as presentation level increased as expected. For individuals with normal hearing, when sentences were presented at 10-dB sensation level (SL), they scored 40%. When the presentation level was increased to 20 dB SL, their score increased to approximately 90%. This translates to a 5% increase in sentence recognition with every 1-dB SL increase from 10 to 20 dB SL. As for individuals with hearing loss, the dB SL increase was even more critical for speech understanding. Based on these results, it could be hypothesized that if a smartphone case caused a reduction in output of 10 dB, one's ability to recognize speech might be reduced by as much as 50%. This could result in a significant communication breakdown and add to one's negative psychosocial experience. Given there are no facial visual cues to supplement phone communication, it is especially important for hearing healthcare professionals to know the impact of waterproof phone cases and counsel patients accordingly.

The purpose of this study was to examine the effects on dB output caused by waterproof smartphone cases that cover the ear-level speaker. Specifically, (1) Are there significant differences in attenuation between having no protective case versus using a protective case on a Samsung Galaxy S7 and Apple iPhone 6s smartphones and (2) Does the attenuation vary as a function of presentation level?

Methods

Equipment

Two smartphone types, Android and iOS, were evaluated. For Android, testing was done with a Samsung Galaxy S7, and for iOS testing was done with two Apple iPhone 6s phones to allow for examination of measurement reliability. Additionally, an Apple iPhone 4 was used to make the test call to the Samsung Galaxy and Apple iPhones. The Samsung Galaxy S7's model number was SM-G930V with operating system Android version 7.0. The two Apple iPhone 6s phones were model number A1688 with operating systems of 11.3 (#1) and 11.2.2 (#2). The volume control on each phone was set to maximum.

The waterproof cases utilized were the LifeProof FRĒ and the OUNNE Shockproof Case. The LifeProof case was available for both the iPhone 6s and the Galaxy S7, but the OUNNE case was available only for the iPhone 6s. Per Life Proof's Web site, for the iPhone 6s, LifeProof FRĒ (item model 77-52563) weighs 1.23 ounces, and will protect the iPhone up to 6.6 feet underwater for 1 hour. ⁶ It “…seals out snow, ice, dirt and dust particles – meets or exceeds IP-68 Ingress Protection Rating.” Ingress protection (IP) ratings are defined in international standard EN 60529. ⁷ They are used to define levels of sealing effectiveness of electrical enclosures against intrusion from foreign bodies and moisture. The first number in the rating represents intrusion, and the second number represents moisture protection. Using the LifeProof FRĒ as an example, the 6 represents “protected against dust that may harm equipment” and 8 represents “protected against temporary immersion.” It is also shockproof, “withstands drops from 6.6 feet – meets or exceeds MIL STD 810F-516.” MIL STD810F is a military standard that aims to define environmental stress levels of equipment. Another one of their selling points is that the case “shields ports, mic + speakers.” ⁸ The LifeProof FRĒ's shell is polycarbonate, with some synthetic rubber and silicone for grip and durability. The screen protector is a transparent polymer. An optical glass covers the camera lens on the back of the phone to protect photo quality. ⁹

The OUNNE (item model 4326917553) weighs 1.76 ounces, is a generic case, much like the LifeProof FRĒ. The manufacturer does not have an official Web site, but rather, they sell on Amazon, an online marketplace. No specific specs are provided and it is not IP or MIL rated. Under “Tips” on the item's listing, it does suggest not putting the phone case in water for a long time at depths more than 6.6 feet. In broken English, it also states, “Making calls: sounds might be lower than expect it, since the waterproof case is design to be tightly shut excluding even sound in and out of the device.” Though not specified, the materials of the OUNNE are expected to be similar to those of the LifeProof.

The Frye FONIX FP35 test box generated the test signal and an AudioScan Verifit1 test box was used to measure the output of the phones. Prior to each data collection session, the Frye FONIX FP35 test box was leveled and the AudioScan Verifit1 was calibrated.

Stimuli

A composite stimulus was presented to the phones to represent a speech-like signal. Per the Frye FONIX manual, “the standard composite signal is composed of 79 different individual frequencies that are generated with a random phase pattern that results in the signal having a crest factor of 20 dB, close to the crest factor of human speech.” ¹⁰ Chasin explains the crest factor as a measure of the difference in decibels between the peaks in a spectrum and the average or root mean square (RMS) value. ¹¹ Typically, the crest factor for speech is approximately 12 dB; the Frye FONIX system has a crest factor of 10 dB. Three stimulus levels—65, 70, and 75 dB SPL (sound pressure level)—were tested for each phone and case condition.

Procedure

The Apple iPhone 4 made an active phone call to the test phone, and then was placed inside the FONIX FP35 test box as shown in Fig. 1 .

Figure 1.

The iPhone 4 inside the Fonix FP35 test box.

The phone being tested (iPhone 6s or Samsung Galaxy S7) was placed in an AudioScan Verifit1 test box and the output of the active phone call made was measured using the test box microphone. The microphone was placed right over the phone's ear-level speaker and secured with Loctite mounting putty, as shown in Fig. 2 . Both test box lids were closed as shown in Fig. 3 and the RMS output of each phone and waterproof case condition was recorded. Measurements were made with and without each waterproof case for each phone. For test–retest reliability, each condition was recreated three times for a total of nine output measurements for each waterproof case.

Figure 2.

The test phone inside the Verifit1 test box with the test box microphone attached to the ear-level speaker.

Figure 3.

The iPhone 4 from inside the Fonix FP35 in an active phone call to the test phone inside the Verifit1 test box.

Results

Output Measurements

The RMS outputs for the Samsung Galaxy, iPhone 6s #1, and iPhone 6s #2 are shown in Figs. 4 , 5 , and 6 , respectively. Overall, the RMS outputs were higher for the Samsung Galaxy S7. For each phone, there was a reduction in output with the waterproof case compared with no case across the three stimulus levels. The reduction ranged from 1.50 to 19.50 dB with the LifeProof case on the iPhone 6s, 9.50 to 27.50 dB with the OUNNE case on the iPhone 6s, and 19.00 to 39.50 dB with the LifeProof case on the Samsung Galaxy S7. For the iPhones, the LifeProof case attenuated less than the OUNNE case.

Figure 4.

Average leveled RMS output values for the Samsung Galaxy S7 with the LifeProof case at all three presentation levels. Error bars represent ± 1 standard deviation. RMS, root mean square.

Figure 5.

Average leveled RMS output values for the Apple iPhone 6s #1 with the LifeProof and OUNNE cases at all three presentation levels. Error bars represent ± 1 standard deviation. RMS, root mean square.

Figure 6.

Average leveled RMS output values, for the Apple iPhone 6s #2 with the LifeProof and OUNNE cases at all three presentation levels. Error bars represent ± 1 standard deviation. RMS, root mean square.

Attenuation

For statistical analysis, the amount of attenuation caused by each case was determined. The differences between the no case and each case condition were found for each phone and at all presentation levels. These differences will be referred to as attenuation. For example, at the 65-dB SPL presentation level, with no case, the average Apple iPhone 6s #1 RMS output level across three measurements was 73.50 dB SPL, and with the LifeProof case the RMS output level was 65.50 dB SPL. Therefore, the difference, or attenuation, that the LifeProof case caused was 8 dB. The attenuation levels, means, and standard deviations for all phone types, case conditions, and the three presentation levels are included in Table 1 . A two-factor ANOVA with replication was performed on each of the Apple iPhone 6s #1 and #2. The independent variables were the two case conditions and the three presentation levels. The dependent variable was the amount of attenuation caused by the protective cases, the LifeProof and OUNNE. For the iPhone 6s #1, a significant effect of case and presentation levels was found, but no significant interaction ( F (1, 48) = 56.91, p  < 0.001 and F (2, 48) = 12.83, p  < 0.00). Post hoc t -tests were performed on the presentation levels. There was a significant effect among all of the presentation level comparisons, 65 versus 70 dB SPL ( t (17) = −4.31), p  < 0.05, 65 versus 75 dB SPL ( t (17) = −5.59, p  < 0.05), and 70 versus 75 dB SPL ( t (17) = −2.76, p  < 0.05).

Table 1. Average Attenuation in dB/± 1 Standard Deviation the LifeProof FRE and OUNNE Case Cause on the Apple iPhone 6s #1 and #2 and the Samsung Galaxy S7 at the Three Presentation Levels.

Phone type	Case type	Presentation level (dB SPL)
		65	70	75
Apple iPhone 6s #1	LifeProof FRĒ	9.22/5.27	13.33/4.34	14.17/4.61
	OUNNE	16.06/1.67	19.72/1.48	23.06/2.11
Apple iPhone 6s #2	LifeProof FRĒ	12.61/5.71	14.67/2.36	14.83/2.11
	OUNNE	16.83/5.15	20.67/4.56	23.33/3.13
Samsung Galaxy S7	LifeProof FRĒ	27.50/3.67	32.56/1.26	37.44/2.04

Abbreviation: SPL, sound pressure level.

The next two-factor ANOVA with replication on the iPhone 6s #2 revealed identical results to those obtained with the iPhone 6s #1. A significant effect of case and presentation levels was found, but no significant interaction, ( F (1, 48) = 31.59, p  < 0.001 and F (2, 48) = 5.35, p  < 0.001). Post hoc t -tests were performed on the presentation levels. Again, there were significant effects among all of the presentation level comparisons: 65 versus 70 dB SPL ( t (17) = −2.23, p  < 0.05), 65 versus 75 dB SPL ( t (17) = −3.04, p  < 0.05), and 70 versus 75 dB SPL ( t (17) = −2.55, p  < 0.05). For the two iPhone 6s phones, the attenuation caused by the OUNNE was significantly greater than the attenuation caused by the LifeProof, regardless of the input level. Additionally, the attenuation caused by both cases increased significantly as the input level increased.

Because there was only one case evaluated for the Samsung Galaxy S7, a one-way ANOVA was performed and showed a significant effect of presentation level ( F (2, 25) = 34.81, p  < 0.001). Post hoc t -tests were performed among the three presentation level comparisons. Significant differences were found among all levels, 65 versus 70 dB SPL ( t (16) = −3.91, p  < 0.05), 65 versus 75 dB SPL ( t (16) = −7.11, p  < 0.05), and 70 versus 65 dB SPL ( t (16) = −6.12, p  < 0.05). Following the same trend as the two iPhone 6s phones, the attenuation caused by the LifeProof case increased significantly as the input level increased. When compared with the absence of a waterproof case, on average, across presentation levels, the LifeProof case attenuated the output 32.50 dB for the Samsung Galaxy S7 and 13.14 dB for the two iPhone 6s phones, and the OUNNE case attenuated the output 19.94 dB for the two iPhone 6s phones.

Discussion

The results illustrate that waterproof smartphone cases do significantly attenuate the output of a smartphone's ear-level speaker during an active phone call when tested using a composite signal. It should be noted that these results were obtained without consideration for the use of amplification. Additional research with behavioral testing of speech recognition as a function of waterproof case and hearing aid use would provide valuable information for counseling patients regarding success with smartphones when cases are used.

Using data from Hood and Poole, ⁵ it can be estimated that the 33-dB attenuation caused by the LifeProof case could result in a complete lack of intelligibility of the signal when used with the Samsung Galaxy S7. When used with the iPhone 6s, the LifeProof case could result in a 60% reduction in speech recognition.

These results, when applied clinically, can facilitate proper and informed counseling approaches by audiologists with their patients, with and without hearing loss, in regard to smartphone use. The attenuation caused by the cases could impact the patient's ability to understand on the phone to a significant degree. A convenient way to avoid this communication challenge for persons who are candidates for amplification is to counsel them on the attenuation the waterproof cases on the intensity of the signal and suggest a different protective case. If having a waterproof case is important to an individual with hearing difficulties, it should be suggested they use a protective case without material over the ear-level speaker and to use the waterproof case when participating in water or snow activities and/or considering using text messaging for communication.

Limitations

This study has several limitations, one of which is a small sample size. There are many other popular smartphones and waterproof cases, and many more models created by the same manufacturers than what was assessed in this study. The findings cannot be generalized to other phone types, model, or manufacturers, or to other phone cases not examined in this study. Another noteworthy limitation is the lack of sensitive test measurements. Clinical equipment, the AudioScan Verifit, was adapted to measure output of a smartphone in a nontraditional way.

Future research should be conducted to examine a broader range of phone types and use a more sensitive microphone and software for measuring the output. Such capabilities could be used to examine if the phone cases change the spectral output/frequency response of the signal.

Lastly, what was found to be significant statistically may or may not translate into functional deficits in human individuals. A study incorporating a behavioral testing component that assesses human performance among the conditions would determine if these differences were clinically significant or not. Assessing a population of individuals with normal hearing and a variety of hearing loss configurations would help identify the groups most impacted by these findings.

Summary

In summary, the presence of a waterproof case on a smartphone significantly attenuated the output of a composite noise signal. Significant differences were found for the amount of attenuation caused by the OUNNE case when compared with the LifeProof case on the two iPhone 6s phones. The amount of attenuation increased as the presentation level increased, across all phone and case types in a nonlinear fashion.