Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to page options Skip directly to site content

So How Accurate Are These Smartphone Sound Measurement Apps?

Posted on by Chucri A. Kardous, MS, PE and Peter B. Shaw, Ph.D.
Figure 1. The SoundMeter app on the iPhone 5 (L) and iPhone 4S (R) compared to ½” Larson-Davis 2559 random incidence type 1 microphone (C).

As of June 2013, 60% of all mobile subscribers use smartphones—that’s more than 140 million devices. Apple iOS and Google Android platforms account for 93% of those devices [Nielsen, 2013]. Smartphone developers now offer many sound measurement applications (apps) using the devices’ built-in microphone (or through an external microphone for more sophisticated applications). The use of smartphone sound measurement apps can have a tremendous and far-reaching impact in the areas of noise research and noise control in the workplace as every smartphone can be potentially turned into a dosimeter or a sound level meter [Maisonneuve et al., 2010; Williams and Sukara, 2013]. However, in order for smartphone apps to gain acceptance in the occupational environment, the apps must meet certain minimal criteria for functionality, accuracy, and relevancy to the users in general and the worker in particular.

Chuck Kardous recording measurements
Video: CAPT Kardous testing mobile sound-meter apps in the lab

NIOSH noise researchers received numerous requests from stakeholders, safety professionals, and the public to address the accuracy of the many sound measurement applications available for smartphones and whether they can be relied upon to provide an accurate assessment of the ambient environment. As a result, we conducted a pilot study to select and characterize the functionality and accuracy of these apps as an initial step in a broader effort to determine whether these apps can be relied on to conduct participatory noise monitoring studies in the workplace [Kardous and Shaw, 2014]. The resulting paper, Evaluation of smartphone sound measurement application, was published in the Journal of the Acoustical Society of America.

We selected and acquired a representative sample of the popular smartphones and tablets on the market as of June 2013. Smartphone apps were selected based on their ability to measure occupationally relevant sound level values. A total of 130 iOS apps were examined and downloaded from the iTunes store*, of those, 10 apps met our selection criteria. A total of 62 Android apps were examined and downloaded, however, only 4 apps partially met our criteria and were selected for additional testing. As a result, a comprehensive experimental design and analysis similar to the iOS devices and apps study above was not possible.

The measurements were conducted in a diffuse sound field at a reverberant noise chamber at the NIOSH acoustics testing laboratory. For our experimental setup, we generated pink noise with a 20Hz ‒ 20kHz frequency range, at levels from 65 dB to 95 dB in 5-dB increments (7 different noise levels. Reference sound level measurements were obtained using a ½-inch Larson-Davis (DePew, NY) model 2559 random incidence microphone. Additionally, a Larson-Davis Model 831 type 1 sound level meter was used to verify sound pressure levels. Smartphones were set up on a stand in the middle of the chamber at a height of 4 feet and approximately 6 inches from the reference microphone as shown in Figure 1.

 

Overall, the results in Figure 2 show that for A-weighted sound level measurements three apps had mean differences within ± 2dBA of the reference measurements. For un-weighted sound level measurements three apps had mean differences within the ± 2 dB of the reference measurement. Since national standards and occupational guidelines specify that type 2 sound measurement instruments have an accuracy of ± 2dBA, some of the above-mentioned apps could potentially be used in the occupational setting, especially if they’re used in conjunction with a type 2 external microphone such as the MicW i436.

Android-based apps lacked the features and functionalities found in iOS apps. This is likely due to the iOS advanced audio capabilities compared to other platforms, the open ecosystem of the Android platform, and having so many different Android device manufacturers using different suppliers and components.

Challenges remain with using smartphones to collect and document noise exposure data. Some of the main issues encountered in recent studies relate to privacy and collection of personal data, sustained motivation to participate in such studies, bad or corrupted data, and mechanisms for storing and accessing such data. Most of these issues are being carefully studied and addressed [Drosatos et al., 2012; Huang et al. 2010].

In conclusion, smartphone sound apps can serve to empower workers and help them make educated decisions about their work environments. They may be useful for industrial hygienists and OS&H managers to make quick spot measurements to determine if noise levels exist in a workplace that can harm workers’ hearing. The ubiquity of smartphones and the availability of these sound measurement apps may also present new research opportunities for occupational hearing scientists and researchers.

Chucri A. Kardous, MS, PE and Peter B. Shaw, Ph.D.

CAPT Kardous is a research engineer in the NIOSH Division of Applied Research and Technology.

Dr. Shaw is a statistician in the NIOSH Division of Applied Research and Technology.

*References to products, services, or apps do not constitute an endorsement by NIOSH or the U.S. government.

References

Drosatos, G., Efraimidis, P. S., Athanasiadis, I. N., D’Hondt, E., & Stevens, M. [2012]. A privacy-preserving cloud computing system for creating participatory noise maps. In Computer Software and Applications Conference (COMPSAC), 2012 IEEE 36th Annual (pp. 581-586). IEEE.

Huang, K. L., Kanhere, S. S., & Hu, W. [2010]. Are you contributing trustworthy data? the case for a reputation system in participatory sensing. In Proceedings of the 13th ACM international conference on Modeling, analysis, and simulation of wireless and mobile systems (pp. 14-22). ACM.

Kardous, C. A., & Shaw, P. B. (2014). Evaluation of smartphone sound measurement applications. The Journal of the Acoustical Society of America, 135(4), EL186-EL192.

Maisonneuve N., Matthias N. [2010]. Participatory noise pollution monitoring using mobile phones. Information Polity, 51-71.

Nielsen [2013]. Mobile Majority: U.S. Smartphone ownership tops 60%. Retrieved June 23, 2013, from http://www.nielsen.com/us/en/newswire/2013/mobile-majority–u-s–smartphone-ownership-tops-60-.html

Williams W. and Sukara Z. [2013]. Simplified noise labelling for plant or equipment used in workplaces. Journal of Health and Safety, Research and Practice, Vol. 5 (2), 18-22.

Posted on by Chucri A. Kardous, MS, PE and Peter B. Shaw, Ph.D.

154 comments on “So How Accurate Are These Smartphone Sound Measurement Apps?”

Comments listed below are posted by individuals not associated with CDC, unless otherwise stated. These comments do not represent the official views of CDC, and CDC does not guarantee that any information posted by individuals on this site is correct, and disclaims any liability for any loss or damage resulting from reliance on any such information. Read more about our comment policy ».

    NIOSH does not recommend nor endorse a particular commercial product. The data from our study “Evaluation of smartphone sound measurement applications” is available in the journal article referenced at http://scitation.aip.org/content/asa/journal/jasa/135/4/10.1121/1.4865269. The discussion section (page 189) highlights two apps — SoundMeter and SPLnFFT — that had the best accuracy over our testing range. There were two other apps — Noise Hunter and NoiSee — that were within ± 2dB of the reference sound source. Once again, note that the data results in the study do not constitute as NIOSH endorsement of any of these products.

    The question of any technology designed to collect data is to ask “now what” or “what do we do with this data?” If these apps, whether for noise or other occupational exposure purposes, will continue to pervade our society, a guidance document to inform public and workers on what to do with the data is appropriate. For compliance purposes, this bears serious examination since compliance may entail expensive fines or other regulatory actions. For general information to prompt workers and the public to employ safe practices and/or hearing protection until detailed analysis with actual noise survey equipment is prudent. NIOSH should seriously consider producing this guidance.

    Thank you for your comment. You are correct that compliance officer measurements will still need to be made with calibrated survey meters, not smartphone apps. Additionally, if an employer is considering large investments in noise control equipment they may wish to employ professionals who have more complex analytical instruments to assure the control designs are efficient and effective. Nevertheless, smartphone apps can serve a useful function as range-finding devices and may help workers make educated decisions such as knowing when to seek help or use hearing protection; and they may be useful for industrial hygienists and OS&H managers to make quick spot measurements when a professional survey meter is not available. Many years ago NIOSH did publish an “Occupational Exposure Sampling Strategy Manual” which is still available on the Internet at: http://www.cdc.gov/niosh/docs/77-173/pdfs/77-173.pdf. It describes the use of range-finding instruments and also strategies for monitoring compliance and enforcement. While the 1977 manual didn’t envision using smartphones, the strategies in that old document still apply for sampling and may guide the use of survey data. NIOSH is in the process of updating the Occupational Sampling Strategies Manual. We will consider your suggestion for updating our guidance on data analysis. NIOSH is also in the process of standing up a virtual center that will focus on guidance documents with respect to direct reading monitors and sensor technology.

    As an industrial hyginist and safety engineer I’ve done research on this field in 2013 as part of my bachelor safety engineer. I’ve witten a thesis with the title ‘smartphone as an Oh&s professional’ (in dutch). Everbody can download it from [http://www.iArbo.nl] OR [http://www.ohs-apps.com]
    Maybe it is possible to translate the thesis with [http://translate.google.com]

    In this thesis have done (1) an exploration of the use of ohs apps under the OHS professionals in the Netherlands and the effect on their work (614 ohs professionals filled in my survey). (2) an exploration of the available apps in the field of Oh&s (3) comparison the use of apps clinical medical field and the ohs field: what are risks OR opportunities…
    During the writing of the thesis I also found extended labtests of smartphone types in different combinations with different noise apps!! In my thesis you Will find the references.
    I have als done à practical comparison test with CO2 sensor (gadget for approx $150) versus à TSI QTRAK (approx. $2000)

    It is definitly opening an new research field and (i believe) an necessity for ohs professionals to learn about the apps risks and possibilties. It has an potential for empowering employees and ohs professionals.
    Questions? hoevers@gmail.com
    With Fr. Grtz Ronald Hoevers

    Thank you for sharing your thoughts and your thesis. We will get in touch if we have any questions.

    This is really a comprehensive article that I have gone through at the moment. I believe that accuracy of the apps depend upon the skill and approach of the mobile app developer and if the app is customized it also depend upon the way things and details explained by the people who need it.

    Thank you for your comment, we agree that the accuracy of some apps highly correlates to the skill and approach of the app developer. Some of these apps have been developed by highly-skilled engineers and programmers who have a great understanding of the principles of acoustics, sound measurement, and applicable noise standards.

    Thanks for your interesting article. In July 2013 I have tested several free apps myself mainly to see how my app SafeNoise was doing compared to other free apps. You can find an article about my test on

    [http://www.geluidnieuws.nl/2013/jul2013/app.html] (only in Dutch).

    In my test I have used 5 samples:
    – pink noise with a sound level of 80 and 100 dB(A)
    – pure tones of 80 Hz (81 dB(A) and 4 kHz (93 dB(A)
    – rockmusic 93 dB(A)
    The setup was straight forward using a stage floor monitor (35 Hz to 16 kHz), a PA amplifier and a laptop with the sound samples. The smartphone and the reference microphone were located close to each other about 1.5 mtr distance from the speaker and pointed directly to the speaker. Due to the short distance between speaker and microphones, there was no influence of the room acoustics (about 65 m3, RTs between 0,5 and 0,2).
    I also checked the influence of the position of the smartphone and the reference microphone (figures 3 and 4 in my article).

    The iOs apps turned out to be better than the Windows or Android apps. Some iOs apps performed within 2 dB(A) of the reference soundlevel meter (MusicSafeCheck and SafeNoise). Did you test those apps as well? If so, how did they do?

    Thank you for sharing your article, it would be greatly beneficial to the readers of this blog if there is an English translation or if you plan to publish it in a peer-reviewed journal. We examined hundreds of apps but selected 10 iOS apps and 4 Android apps for additional testing. None of the Windows-based apps met our selection criteria. See our journal article for our selection criteria of the apps we tested and their performance.

    I was surprised to find that the microphones used in the Apple phones have a fairly flat frequency response, as I thought that all cell phone mics were intended to capture speech and therefore had poor responses at both the low and high end of the frequency spectrum. That means that mics used in Android devices, even when the apps have been properly calibrated, can give a fairly accurate sound level for sounds in the speech frequency range (500 Hz to 2000 Hz range), but can give very inaccurate results when recording other sounds with low and high frequency content, such as from music and diesel engines.

    The microphones themselves exhibit a good flat frequency response up to around 10-12 kHz depending on the type of MEMS microphone used in each device, the smartphone manufacturers introduce speech and noise cancelling filters into the system for normal communication needs. However, starting with iOS 6, Apple allowed developers to bypass such filters. Android developers must still work around such filters during the development of their apps.

    As far as microphones used in Android devices, they’re usually similar to those found in Apple devices, almost all smartphone manufacturers rely on few suppliers of MEMS microphones. The difference in performance and accuracy across devices and platforms are not necessarily related to the microphones but rather the set of software and drivers that handle and process the audio signal. Apple offers its developers access to their digital audio infrastructure, called Core Audio, which is tightly integrated with the iOS. No such infrastructure exists for other platforms such as Android and Windows as they continue to rely on third-party drivers and that can introduce latency and affect the performance of the apps.

    Thanks for this investigation.
    Steady broadband noise test signal: the test should be extended with offering bursts (in free field!) and signals with high (but realistic) crest factors.
    The test should also be extend with performance at different temperatures for example. Some other factor maybe added, for example if it has a fair high frequency response (think about compressed air noise).

    Then, if you clearly inform the user about all limitations, selected apps could be very well useful. Of course no one should expect the “full performace” of such apps. But it will be used by a large audience with little knowledge about the subject; then its important that they know were it goes wrong.

    Thank you for your feedback, we will take your input into consideration if we decide to expand our study.

    I cannot thank you enough for the blog article.Really looking forward to read more. Really Cool.
    schalke trikot

    OK, So I’m far from an intellectual and honestly just barley graduated High School. Most of what’s said here is Greek To Me. I work in a night club doing as best as I can for My Family with what I have. I’ve noticed several of the older people there who have admitted to being in clubs for a long time have difficulties in hearing. What I’m trying to do is find a device / app that I can turn or / activate that can give me a”DECENT” reading and tell me if I could be wearing hearing protection or not ? I understand that some people or business’s are not allowed to say choose this one over this one, but as a person who DOES NOT want to lose his hearing and can not quit right now, could someone PLEASE put in PLAIN ENGLISH the good, the bad or the indifferent. I have an older iphone 4 and really want to know. Thank You. Dan

    We appreciate your comment and sincere request. Please note that NIOSH does not recommend nor endorse a particular commercial product and also note that this was a pilot study so there may be other apps out there that may be as accurate as the ones we tested. The results of our study showed that the SPLnFFT app from Fabien Lefebvre and SoundMeter app from Faber Acoustical had the best accuracy, followed by Noise Hunter and NoiSee. All of these apps are available through the iTunes app store. Take a look at the description and screen layout of each and choose an app that you feel most comfortable using. Any of these apps should be adequate to do the job (The apps pricing range from $0.99 for NoiSee to $19.99/$99.99 for SoundMeter).

    The most important thing is to get 3-5 readouts over different periods of times so you can get a better understanding of the noise environment at your club and find out what the overall average noise levels. To make sense of the noise level readouts you get from a particular app, note that the NIOSH recommended exposure over an 8-hour workday is 85 decibels, A-weighted (dBA), and for every 3-dB increase in that level, you cut exposure time in half, so if your reading is 88 dBA, , you should only be exposed to that level of noise for up to 4 hours (per day), at 91 dBA, exposure time is cut to 2 hours, and so on. See Table 1-1 in this document http://www.cdc.gov/niosh/docs/98-126/pdfs/98-126.pdf and try to match the noise level readout to the recommended exposure duration. Remember, these apps are only meant to give you a rough idea about your environment, they do not replace the need for a professional noise assessment. But if your readouts exceed our recommended limit of 85 dBA on a consistent basis, then it might be worthwhile to share that information with the club management and see if they can address it. There are a variety of solutions available to an employer, from reducing and controlling the noise levels, to administrative controls such as limiting time spent in the area with the excessive noise and taking breaks, or if those are not possible, offering hearing protection. There are many different kinds of hearing protectors available these days, such as the “musicians’ earplugs” that attenuate noise uniformly across all frequencies without distorting speech and music. Based on your description of your job, the musicians’ type of protectors would seem to be ideal for your situation.

    Very informative article. I run Techogeek which is based on Smartphone apps and reviews and i will definately write something for you

    This article Seems to be nice. It is really informative as well as well written. Also looking forward to read another article like this. Once again thanking you for such an article..

    I believe that accuracy of the apps depend upon the skill and approach of the mobile app developer and if the app is customized it also depend upon the way things and details explained by the people who need it.
    Super

    Chanced upon this study back in May and proceed to ‘mix and match’ the setup with the SoundMeter app coupled with the i436 and used it for area noise measurements at some workplaces in Singapore. The results, when compared with a calibrated SLM type II are pretty close.

    Thank you for your comment, please share with our readers some of your findings. We are in the process of conducting a study with several external microphones.

    It is incredible that a smartphone can do so much these days. I believe that a smartphone will replace the laptop one day

    Peter Gajetto

    I cannot thank you enough for the blog article.Really looking forward to read more. Really Cool.
    schalke trikot

    Thanks for your interesting article. However I would like to understand something. 1.Why you don’t speak about the hudge variability that’s posible to see in the figure 2 ?
    2.The App was more accurate at low level or High Level ?

    By the way, that a good investigation,

    Best regards.

    Good questions, Pierre. We do discuss the variability in our journal article and also in our more detailed report here: http://www.cdc.gov/niosh/surveyreports/pdfs/349-12a.pdf. Please note that such variability is expected since we were testing the same apps with several different iOS devices (iPhone 3GS, iPhone 4S, iPhone 5, and iPads). We would likely see similar variability if we would test several dosimeters, for instance, from several manufacturers over that same test range. And that’s why we decided to use the means of the differences between each of the apps (and each of the devices we tested) against the reference sound levels as a measure of accuracy. As for your second question, each app behaved a little differently, but we saw a little more variability at the low test level of 65 dB compared to the higher test levels of 90-95dB.

    The four android apps that partially met our criteria are:
    SPL Meter from AudioControl
    decibel Pro from BSB Mobile Solutions
    db Sound Meter from Darren Gates
    Noise Meter from JINASYS.

    More information can be found on page 7 at http://www.cdc.gov/niosh/surveyreports/pdfs/349-12a.pdf

    References to products or services do not constitute an endorsement by NIOSH or the U.S. government.

    Interesting article to read. I never thought my smart has such potential. By the way sound apps may get improved with research. Hope you article gets noticed by the sound app developers..

    An interesting study, thanks for posting. It’s worth noting that care should be taken when evaluating sound with a lot of low frequency content – even when using good plug-in microphones such as the MicW i436. The signal still goes through a ~100Hz low frequency filter as far as I can tell.

    Good point, we do note in our study that with the introduction of iOS 6, Apple allows developers of audio apps to bypass the speech filters. Though not all developers have taken advantage of this feature.

    Great research on sound apps of Smartphone!! I guess for the first time someone has taken up such an complex experiment.

    Having read this I thought it was rather informative. I appreciate you finding the time and energy to put this content together. I once again find myself spending a lot of time both reading and posting comments. But so what, it was still worth it!

    Whatever you say recent study shows that
    Google android is so much afraid of apple’s upcoming new iphone. Future is about apps and not about the sound apps.!!

    Hi there,

    Thanks for your article, I found it very educational and useful. It’s nice to read an article like that with a high quality of information and true, honest review. Thanks again for your information.

    Các mặt khắp nơi của điện thoại thông minh và sự sẵn có của các ứng dụng đo lường âm thanh cũng có thể trình bày những cơ hội mới cho các nhà khoa học thính giác nghề nghiệp và các nhà nghiên cứu.

    English translation:
    The ubiquity of smartphones and the availability of sound measurement applications may also present new opportunities for scientists and occupational hearing researchers.

    Excellent sound applications work on smartphones. Science never ceases to amaze. Great article.
    Greetings and congratulations.

    I am an ear specialist. I believe that everyone should download a sound meter on their smart phone, and urge others to do the same. This would help to avoid the hearing loss patients I see daily.
    Can you help? I need a sound meter app for my web site and I am trying to find one that is available.
    drgrossan@yahoo.com

    The use of smart phones is ever expanding. I am a civil engineering student and have recently begun to find ways to use our phones to improve our ability to not only increase safety, but also efficiency. This research has encouraged me to review more apps, and see how to program new ones that meet my specific needs.

    Thanks for sharing such great information with us. I really like your thoughts about sound apps of Smartphone’s. It is really a good research. Keep posting such more post..

    This page is a valuable resource! I would like to share an article that was recently published in Noise & Health journal. It also evaluated the accuracy of sound level measurement apps. Similar results were noted for one of the apps that was measured in the CDC study; however, some of the apps did not record results as accurately in our study as they did in CDC study.

    Read the new article here: http://www.noiseandhealth.org/text.asp?2014/16/72/251/140495

    Thank you for taking the time to read the article and I’m glad you enjoyed it. It’s interesting to see repeatable data for SoundMeter app.

    Have you been able to test accuracy of the apps in louder environments yet?

    We chose 95 dB as the upper limit for our study because of our interest in typical occupational exposures. Our system is capable of generating 110-115 dB but we haven’t tried to push its limits. We are currently exploring the accuracy of apps with external mics, we have used them with acoustic calibrators generating 114 dB at 1000 Hz and 124 dB at 250 Hz and the main apps from our previous study all performed well were able to measure those levels without any issues. We will be posting updates on our twitter account @NIOSHNoise soon.

    I have read so many articles but I would say this is the one of the best article in sound measurement apps. thank you for sharing with us.

    I really appreciate your work. And all that I want to say is Keep sharing such useful information for us. This is the best article on Sound measurement Apps that I ever read ! Can you write a similar article on other type of smartphones apps ? If done, it would be nice work because people like us are waiting for such articles from high authority sites like yours. Anyway thanks a lot.

    Thanks

    As a person with hearing loss, I agree with Dr. Grossan that everyone should have a way to know when the noise level around them is dangerous. We live in a very noisy world. I think there are OSHA regulations for noise in the workplace, but it is a shame that individuals have to have personal DB meters to know if the noise level around them is safe. Smoking has been banned in most public places because second hand smoke is dangerous. Why do we allow noise levels in restaurants , bars, theaters, airplanes, and other public places that can damage hearing? I found this web site, because after being at a party last night and a restaurant this morning that were both painfully noisy, I decided that I did need a personal DB meter so I don’t lose what little hearing I have left. I also want one so I can walk up to a movie theater manager and say “Here is the DB level in your I-Max. It’s too loud. Turn down the volume or give me my money back.”

    Your posts is really helpful for me.Thanks for your wonderful post. I am very happy to read your post. It is really very helpful for us and I have gathered some important information from this blog.

    Thank you very much for sharing this useful information. I was doing a project and for that I was looking for related information. Some of the points are very useful. Do share some more material if you have.

    As a safety manager on a military installation, I appreciate these types of studies. I have used sound measurement iOS apps but never really had too much faith in their accuracy over our professional noise metering equipment. I tend to use them just to get a preliminary reading if I think there is a problem. I also own a site where I have conducted reviews of some of these apps and the information here would be a great addition to what I have previously encountered.

    We use a kiosk that has a hearing test app on it. It’s a tone test. We use it on a Microsoft RT tablet. It works really well. I hope that sound measurements continue to improve as well to help workers understand the noise around them.

    Thanks so very much for taking your time to create this very useful and informative site. I have learned a lot from your site. Thanks!!

    Mr. Stokes,

    I am very interested in how these apps perform in real-world environments compared to calibrated sound level meters. Is there a place where we can see what apps you have compared and their accuracy?

    Most of the apps I believe have the possibility to adjust the gain in order to be exactly calibrated at a specific frequency. Typically 94dB at 1000Hz. But in the posting here I can not see if this is done before the tests. Also it is often a problem at low levels that electric noise in smartphone apps using the built in microphones is around 30-35 dB(A). If you do not know the actual low limit at relevant frequncies then bevare. But for levels above say 45 dB(A) use it for a survey guide !

    Thank you for your comment, Birger. We did not calibrate the apps as we could not establish a viable and repeatable protocol for accurately calibrating the internal mics. You simply cannot place a calibrator on the opening of the internal mic, and that’s the reason we decided to expand our study to examine the accuracy of sound measurement apps using external mics which allowed us to use acoustic calibrators as they’re intended to be used. See photos here:
    https://twitter.com/NIOSHNoise/status/522456711882629120. We will be publishing results soon, stay tuned.

    nice article for me.. i hope the developer read it n do something better.. i think i will share this article to my group, is it okay?

    maybe it can help you to compare the smartphone nowadays

    Thank you for interest Randy, please feel free to spread the word to your group and we’re always looking for feedback and suggestions.

    While the paper was very interesting, it does not do enough to caution the potential user that the resulting measurements do not come near to meeting International IEC or even local US ANSI measurement standards, nor to accurately measuring many “real world” noises sources, however close to the “truth” it seems on simple sinusoidal noise sources.

    Even the most reputable manufacturers of professional meters, such as B &K, Cirrus Research, Rion, Pulsar or 3M-Quest have not found it easy to have their sound level meters accepted and approved by a standards body such as the PTB in Germany; in other words to meet the legally required measurement standards of accuracy is hard to do – indeed, some mid-range sound level meter manufacturers have never managed to get their meters approved. Low frequency has been mentioned as a problem as has calibration and non steady noises sources. If calibration is out and a meter app reads just 3dB too low, the exposure to noise damage risk will potentially be DOUBLED. Similarly if the noise is impulsive with a high LZpeak content, the app may well read 6dB low, so the worker is potentially exposed to FOUR times the safe level. Now add in the potential error for the microphone’s imperfect omnidirectional response as well as temperature, humidity and vibration issues, using the app would seem more like a “goodish guess”. It seems to follow that these apps should NEVER be used as hearing damage risk tools, but just as a very crude indicator, for which task they seem adequate.

    Thank you Dudley, you bring up some great points in your comment. One of the main reason we embarked on such a study, we had stakeholders and industrial hygienists ask us whether such apps can be used to make assessment of noise exposures in the workplace. We continue to investigate the viability of using such apps when adding a type 2 external microphones, that are supposed to have flat frequency responses from 20 Hz – 20 kHz. Our preliminary results showed even greater accuracy of such apps when used with external mics. Impulse noise assessment is a tricky issue by itself because even type 1 sound level meters are often not capable of measuring impulses that exceed the dynamic range of the meter. NIOSH has published extensively on the issues with impulse noise measurements. For example, see:
    http://www.cdc.gov/niosh/surveyreports/pdfs/349-11a.pdf

    http://www.cdc.gov/niosh/hhe/reports/pdfs/2013-0124-3208.pdf

    Noted Chuck. Clearly, whatever the range of an instrument there will ALWAYS be some peak that may exceed the meter’s linear range – that is why these matters are specified in IEC 61672 so it is not relevant is it? As a matter of fact, most B & K and Cirrus microphones – two types fitted to many commercial units – have a linear range from about 16dB(A) up to 153 dB. One important point is that IEC 61672 mandates a Class 2 microphone ONLY up to 8kHz (see IEC 61672 table 2). Even a Class 1 is only fully specified up to 16kHz. At 20kHz the tolerances are +6dB and – infinity (+4dB and -infinity for Class 1). I am aware that ANSI standards have different limits, even though very few commercial units in truth meet the ANSI requirements. In Europe, meters have to be fully tested, but in the USA as far as I am aware, such specification claims do not have to be substantiated by pattern approval and so may not be trustworthy. The only apps we have tested did not even meet the rms test and in truth are little better than $30 ‘toys’ – wonderful for setting up hi-fi and great for giving a scale order of the noise levels, but absolutely not suitable for noise measurement. I feel you would be doing people at hearing damage risk a great dis-service by supporting the use of phone apps.

    noise measurement is very useful application is used for a student and worker who wants to succeed in his job
    thanks for your information…nice job

    nice article for me.. i hope the developer read it n do something better.. i think i will share this article to my group, is it okay?

    Lots of deserved “attaboy” comments obscure a few especially useful comments and responses. Perhaps the authors could add an addendum that references a few of these or even extends the blog post. For example, I could not expand the images enough to read many of the app names so the additional information buried in responses that identified the four best performing apps was very useful. (Some recommended apps were quite inexpensive at $1, $4, or $6). I also found links such as these quite useful:
    http://www.cdc.gov/niosh/surveyreports/pdfs/349-12a.pdf (where I can read the box graph but was overwhelmed by the detail spread across 37 pages)
    http://scitation.aip.org/content/asa/journal/jasa/135/4/10.1121/1.4865269 (shorter published version)
    —–
    http://www.noiseandhealth.org/article.asp?issn=1463-1741;year=2014;volume=16;issue=72;spage=251;epage=256;aulast=Nast
    http://www.safetyawakenings.com/safety-app-of-the-week-42/ (April 22, 2013 – “30 iPad Sound Measurement Apps Reviewed” – be sure to read the updates like the note on higher db levels)
    http://www.safetyawakenings.com/safety-app-of-the-week-7/ (May 19k 2014)

    In conclusion, “attaboy” and a positive “my tax dollars at work”!

    Many smartphone developers have built measurement apps that can potentially turn the device into a dosimeter or sound-level meter. thanks

    We live in a noisy world. It is well documented that sound levels over 85dB can cause hearing loss and tinnitus. This article clearly demonstrates the relative accuracy of the smart phone decibel meter to measure that noise. For more information get a copy of my new book “Hearing Loss Facts and Fiction – 7 Secrets to Better Hearing”

    My trucks bass from one 12″ subwoofer peaks most decibel meters and sound meter apps on smartphones at 90db they don’t register anything higher in my experience

    Based on our measurements of some of the apps we tested, we were able to measure sound levels as high as 120-25 dB SPL without any issues.

    I could not expand the images enough to read many of the app names so the additional information buried in responses that identified the four best performing apps was very useful.

    Android and Apple smartphones have won a kind of sweepstakes leaving other platforms such as Windows Mobile and Palm OS in the dust. Functional used smartphones based on the less favored platforms are cheap enough — often under $10 U.S. — to beg the question whether lemonade can be made from these lemons? What about longer term noise measurement? As good as iOS and Android noise measurement apps have become, you still wouldn’t want to deploy such a device in a public or semi-public place unattended. It very likely would get stolen.

    If long-term (24 hours to one week) noise monitoring could be done with less desirable and cheaply replaceable devices the picture changes. Can an appropriate hardware/operating system platform — cheap, favorable tradeoff of processing capability and battery life, ability to bypass or compensate for speech-oriented frequency shaping, a mic that can handle 100+ dB, accommodating flash memory cards — be identified and then software development be supported? This probably could be launched for something like $10,000 and seems enormously cost-beneficial in terms of making longer-term noise monitoring much more widely doable.

    The software/hardware only needs to capture .WAV files, for post-processing into a number of noise statistics over time, and for extracting clips of incidents of interest. 22,000 16-bit samples per second scales up to just under 28GB per week, comfortably within limits of removable flash memory cards that most smartphones can use. For shorter time periods, higher sampling rates are perfectly feasible, if only to validate the expectation that noise at the highest audible frequencies is seldom very high energy.

    Thank you David, you raise some very interesting and exciting ideas. I believe some citizen science projects have done just that (albeit on a small scale) – used cheaper smartphones and developed a basic sound measurement app, specific for those devices. Some other projects have developed their own cheaper “noise measurement” boxes for environmental noise monitoring. But I agree, there’s great potential to acquire a large number of these “cheaper” devices and pass them along to university students or citizen scientists to conduct a large surveillance effort. The idea of smartphone sound measurement apps is more about personal awareness and empowerment since most people carry one everywhere they go.

    As for NIOSH, our mission is focused on occupational noise exposure and we hope that this effort would serve to empower workers to use their smartphones to become better aware of their noise environment and take actions to reduce their personal exposure to noise.

    The Chinese smartphone ever give a better result.
    There are already many prices and models with increasingly better qualities

    The issue is not the origin of the smartphone or its manufacturer as much as the operating system, and the availability of reliable apps on that operating system. We have tested several Android and Windows-based smartphones, but the apps available were not as advanced or fully-featured as the apps available on iOS. See discussion above about our findings.

    There are a number of design choices that can help or hinder repurposing. For example:
    – Equalization may de-emphasize frequencies outside normal voice range. For noise measurement purposes it should be possible to turn this off.
    – Gain should be adjustable so a phone’s dynamic range can shifted to suit an expected range of environmental noise levels, however AGC needs to be turned off so relationships between noise dB and digitized quantities are known.
    – Being able to save uncompressed .WAV files, 16 or 24 bits per sample, sample rates variable up to 22,050 or 44,100 per second. Compressed representations such as ADPCM or MP3 throw away too much information.
    – Operating systems should not block repurposing.
    – Batteries and flash storage media should be easily replaceable and upgradeable in capacity.
    – USB ports should be reprogrammable into host mode.
    – There may be further requirements to support repurposing beyond noise measurement, e.g. measuring vibration, radiation, light intensity, temperature, pollution.

    Why not make these accomodations mandatory through the FCC, just as radiation, interference, telemarketing and privacy are already regulated? Making phones more friendly to repurposing can promote a number of societal benefits.

    Thank you Brett, decibel meter pro was one of the least accurate in our testing (difference of -13.2 dBA), I am surprised you gave it “5 stars”, what did you base your evaluation on?

    Apple did a great job with iPhones and their other iOS devices in most respects. Android may be pretty good too. Nevertheless there are a lots of other smartphones out there. Given the gravitation of users toward these two winning O/Ss plus mobile phones being on short cycles of obsolescence and replacement, there are an awful lot of still-functioning devices that are roadkill on the way to landfills or incredibly unsustainable recycling in China. That is unless they can be repurposed.

    Great and useful information.

    As a science teacher I was surfing the net for apps to use in the classroom and share amongst the pupils. Both iOS and Android (I use Android myself). As a Dane, I found and shared the Swedish “Buller” app. – published by the occupational health ministry I think -, but was curious as to how well it performed. You haven’t tested it apparently, but your survey is gold anyway!

    Now I know that relatively low but constant levels of both infra- and ultra-sounds – mostly from machinery – can permanently damage hearing, but are there any apps available that can measure infra- and ultra-sound levels at all?

    We actually tested the Buller app from the Swedish Work Environment Authority and though it was well-designed and easy to operate, it did not meet our accuracy criterion. See page 30 of our more detailed report: http://www.cdc.gov/niosh/surveyreports/pdfs/349-12a.pdf.

    As for ultra sounds and infra sounds, the issue is not the apps but the built-in microphones used on most smartphones which have a limited frequency response and introduce distortions at very low and high frequencies. There are external microphones that can be used with smartphones but even those have are limited especially below 20 Hz. For such measurements, you will need a high-end sound level meter and a special attention needs to be made for the selection of the appropriate microphone.

    Mr. Enevoldsen – Electret microphones have good response down into the infrasound range and iOS 7 or higher can turn off the equalization that favors voice frequencies for phone purposes, so it’s likely an app could be written if one doesn’t already exist. Whether this is also feasible for Androids is harder to judge. Apparently there is a lot of variation in the hardware, brand to brand and perhaps model to model.

    Ultrasound measurement would normally be limited by maximum sample rates, 44,100 or 48,000 Hz. Even at 48,000 there would need to be a steep filter blocking frequencies above 21 or 22 kHz to avoid aliasing. Nevertheless, if noise is random you could look at the highest frequencies that can be measured. If there is a substantial amount of energy there, chances are very good that there is also sound energy at higher frequencies out the device’s range.

    As a DIY experiment, perhaps you could build an oscillator with selectable output about 10 kHz apart, through the entire band of interest. If you got a microphone capable of responding in that range, you could amplify that signal with a filter to cut out the audio band, then mix it with oscillator output. Input that into a MIC jack and run an app with spectrum analysis (e.g. SPLnFFT). Then there should be peaks at difference frequencies. For example if there is ultrasound at 31 kHz and you mixed it with 30 kHz, you should see a spike at 1 kHz.

    Perhaps you could build an ultrasound mic and the mixer circuitry into an Altoids tin.

    Thank you for this great work. Is it safe to assume that the accuracy would improve on later iPhones, such as the iPhone6?

    It’s possible, the main issue is in the phones’ built-in microphones and the ability to calibrate to a known sound level. We just completed a series of experiments using iPhone 6s and the measurements are very similar to our earlier results, however, when used with external microphones, the results show much improved accuracy. We’ll report on our new study shortly.

    yes accuracy of the measurement depend on the microphone (also effect opening and the dust cover) . if you dont consider developing a calibrating method ,this is nonsense

    br.

    Thank you for your comment, there are few new studies on methods for calibrating smartphones for noise monitoring and measurements (e.g., Dumoulin et al., 2013 http://scitation.aip.org/content/asa/journal/poma/19/1/10.1121/1.4800063; and Robinson and Tingay, 2014 http://www.acoustics.asn.au/conference_proceedings/INTERNOISE2014/papers/p565.pdf). We just completed our own evaluation of smartphone sound meter apps using external microphones that can be calibrated using standard acoustic calibrators. We will publish our results shortly, we are providing updates on our twitter account @NIOSHNoise (https://twitter.com/NIOSHNoise).

    That’s better than I would have thought, and in some cases on par with very low cost (sub-$50) dedicated SPL meters.
    I have a side business that does audiovisual services, and I’ve been meaning to get the other folks in the crew to test their phone with one of these apps against my SPL meter (which is a ~$80 unit that I believe is pretty accurate in most circumstances), so they’d have a ballpark idea of how loud they’re running the gear (there are noise ordinances that we sometimes have to take responsibility for following).
    I’d be curious how well a phone could act as an RTA, as well. I used to use my Amiga to do analysis (non-realtime, but reasonably useful in a time when having a high resolution spectrum analyzer was amazing), and standard PCs have been able to act as RTAs for at least 15 years, so I’m absolutely certain the processing power needed is there in modern phones and tablets, but the mic and audio interface would be a limiting factor.

    This is an excellent idea, Eva – to allow other crew members to use your SPL meter to calibrate their apps (since calibration is one of the most important things to achieve better accuracy).

    As for RTA, you’re correct, the processing power in these devices allow for some very complex analysis. Two of the apps we tested above offer RTA as “in-app” purchases or as part of larger suite of apps. For such analysis, the use of a “professional grade” external microphone (that can be calibrated using acoustical calibrators) would be crucial to obtain quality results.

    Noise cancelling earbuds are able to measure the ambient noise level that if they have a companion app. The sensitivity of mic integrated in earbuds is greater than one that fit to smartphones.

    Thanks Ted, that’s an interesting application for smartphone apps. I believe noise cancelling earbuds use MEMS mics as well (not sure if similar to smartphone MEMS mics or not) but if you have additional information on their performance and their sensitivity, please share it with our readers. The main advantage for using an external mic is the ability to calibrate, not sure how you can calibrate an “earbud” mic?

    Hi there, thank you for good post, veryinformative. Thanks for sharing, keep up the good work, Thank you, regards.

    Great article and study. Question: how would I measure the sound level inside of a hearing protection muff using one of these apps with an iPhone?

    I wish to determine the approximate amount of sound attenuation that a muff provides. I can use the MicW i436 with an iPhone to measure the ambient noise, but is there another mic that can be used inside of the hearing protection muff for a second reading?

    Thanks for any suggestions.

    Thank you for your interesting question, Richard. I have not thought of such an application, though I do not think these apps and smartphones are suitable for these types of measurements. There are specific standards to measure the hearing protection attenuation (Such as ANSI S12.6). At NIOSH, we typically use an artificial head fixture such as GRAS 45CB (http://www.gras.dk/45cb.html) when we need to evaluate the attenuation of hearing protector devices. You may find this article helpful: http://multimedia.3m.com/mws/media/893207O/preferred-methods-for-measuring-hearing-protector-attenuation.pdf?fn=10%20Preferred%20Methods%20for%20Measuri. If you have a specific earmuff in mind, I would recommend that you consult our hearing protector compendium at http://www.cdc.gov/niosh/topics/noise/hpdcomp/ and use the methods outlined to obtain the true attenuation of protectors.

    However, you may be able to get a very rough approximation of the noise reduction of an earmuff by using a MicW or some other mic (see the Etyomtic Research in-ear mics) with an extension cable running under the foam cup of the earmuff. I would recommend using two calibrated mics and devices unless you can generate the exact ambient sound field when you conduct your measurement under the earmuff. You will also need to secure the mic at the entrance of the ear canal with some form of tape without compromising the seal of the hearing protector too much. I would be interested, as I’m sure some of our readers, to find out more about the specifics of your measurements and your eventual findings.

    Indeed, even the most respectable producers of expert meters, for example, B &K, Cirrus Research, Rion, Pulsar or 3M-Quest have not thought that it was anything but difficult to have their sound level meters acknowledged and endorsed by a benchmarks body, for example, the PTB in Germany; at the end of the day to meet the lawfully required estimation principles of exactness is difficult to do – in fact, some mid-reach sound level meter makers have never figured out how to get their meters affirmed. Low recurrence has been specified as an issue as has adjustment and non consistent clamors sources. On the off chance that alignment is out and a meter application peruses only 3dB too low, the presentation to clamor harm danger will conceivably be DOUBLED. Essentially if the commotion is indiscreet with a high LZpeak content, the application may well read 6dB low, so the specialist is conceivably presented to FOUR times the sheltered level. Presently include the potential mistake for the amplifier’s defective omnidirectional reaction and also temperature, stickiness and vibration issues, utilizing the application would appear to be more like a “goodish theory”. Follow can’t help thinking that these applications ought to NEVER be utilized as listening to harm danger instruments, yet pretty much as an exceptionally unrefined marker, for which assignment they appear to be sufficient.

    The product/equipment just needs to catch .WAV documents, for post-handling into various commotion insights after some time, and for extricating clasps of episodes of hobby. 22,000 16-bit tests every second scales up to just shy of 28GB every week, serenely inside of cutoff points of removable glimmer memory cards that most cell phones can utilize. For shorter time periods, higher inspecting rates are flawlessly plausible, if just to accept the desire that clamor at the most noteworthy capable of being heard frequencies is rarely high vitality.

    I’m really glad I’ve came across this informative post, this really helped me out quite a bit, by answering all my questions properly, hopefully the site will grow bigger and more popular as ever it was. Keep up the good work and I’m sure I’ll visit you up someday soon for more great articles.

    Sound cancelling earbuds are able to calculate the actual normal noises amount if they’ve already a associate app. The level of responsiveness regarding microphone stand bundled inside earbuds is actually higher than one who match to help smartphones..

    Thank you for your informative paper and ensuing discussion.

    My understanding of a ‘noise dosimeter’ is an instrument which measures noise dose; ie as well as measuring the sound level, it integrates the sound level with the time duration of exposure to the sound. The noise dose can then be compared with the recommended safe level of 85 dB(A) for eight hours. The apps which I have seen are noise meters despite being labelled ‘dosimeters’.

    Do you know of an app that actually measures noise dose?

    Of all the apps that we tested, 4 apps met our test criteria, and of those, NoiSee from EA LAB provides a Dose calculation as an included feature of their app. SoundMeter from Faber Acoustical provides an upgrade option for Dose (called %Dose). There may have been other apps that have been introduced recently that offer such features though.

    We do agree that there’s a slew of apps that provide sound level measurements but no averaging or integrating noise exposure features such as TWA or Dose. This is one of several reasons that NIOSH decided to release its own app, developed in collaboration with EA LAB, and it will include all of the typical dosimetry measures such as LeqA, TWA, and Dose, etc. Look for its release soon.

    Are you afraid of the implication on a vendor using your work as a “NIOSH APPROVED” SLM, because you are flawed, this is potentially harmful, by giving False Negative readings, and someone could be harmed by thinking they are not being harmfully exposed.

    No, we are not “…afraid of the implication on a vendor using your work as a “NIOSH APPROVED” SLM, because [we] are flawed.” We do not want any vendor implying endorsement by the U.S. Government (i.e., NIOSH, CDC, or HHS) because that would be a potential violation of the Lanham Act, 15 U.S.C. § 1125, or Section 5 of the Federal Trade Commission Act, 15 U.S.C. § 45(a), as well as other federal laws. Our research revealed that certain apps performed better than others on specific measures. Identifying this does not imply an endorsement of one product over another. This may be a good time to reiterate what we noted in the blog and in our published paper: “References to products, services, or apps do not constitute an endorsement by NIOSH or the U.S. government.”

    Have any Android apps come out since this article was published that have the ±2 dB accuracy required for the type 2 sensors? I teach a Sound Engineering course in Colombia and many of my students can’t afford an iPhone. Thanks in advance for any suggestions.

    Hello Felipe, there have been very few newer sound meter apps on Android but we have not evaluated them. As mentioned in our article, we continue to see the same issues with the Android-based apps – fragmented marketplace for hardware devices, lack of uniformity in audio integration between the different manufacturers, and most importantly, lack of support from the developers community.

    For purposes such as teaching a course, I would focus on precision of measurements instead of accuracy. One of the major issues with Android devices is that a very small percentage run the latest version of the OS. I suggest you have the students download a certain app (see table 9 in our expanded report http://www.cdc.gov/niosh/surveyreports/pdfs/349-12a.pdf), ensure that they all have the same version (latest) of the Android OS, and see if you can use a single sound source to calibrate those phones, collectively, to a known sound level.

    You bring up a good point regarding access, so I invite our readers to contribute to this conversation if they know of specific newer Android apps or ideas that could advance noise measurements on Android devices.

    We will get the word out using a variety of mechanisms when the app is ready for public release. You can
    keep an eye on our @NIOSHNoise twitter account and watch for a new blog.

    Only one word: Thanks for sharing this remarkable research.

    There are many noisy places here, and it’d be appropriate for the residents to become aware of the dangers they expose themselves to daily – especially through something as mundane as their smartphones!

    Great article¡

    I have a question: Have you thought of evaluating the accuracy of the microphones / APPs in outdoor environments.
    Andwhat happens with the influence of wind on the measurements?

    Great question Alvaro, this is something that we have considered – how to translate our lab-based measurements into real-life measurements. We are collaborating with several researchers to evaluate the performance of smartphones to conduct actual field measurements. See recent studies by Murphy and King http://www.sciencedirect.com/science/article/pii/S0003682X15003667 and Ibekwe et al. http://www.tandfonline.com/doi/abs/10.1080/15459624.2015.1093134 to give you a perspective on the performance of smartphone microphones in outdoor environments.

    I actually used one of these apps to help decide the purchase of my last car. I used the phone to measure the noise rating on several vehicles at 70 MPH, normal highway driving conditions. I have a long commute from Morgantown to Pittsburgh so a quiet cabin is very important to me. I am glad to see that my investigation actually had some validity.

    Thanks Michael, would you please share with our readers additional details about your specific experience as some may find it interesting? I think this is a unique use for such apps that we never considered – the contribution of a long commute to a person’s overall daily noise exposure!

    Thank you for the feedback. My experiment found its origin when I began to notice that when I would turn my car on the radio would be blasting. The road noise was so loud that I didn’t realize how loud the stereo was playing from the day’s commute. I began to wonder what the long term effects of all this road noise would be on my hearing. I downloaded Decibel 10th Professional Noise Meter, free from the app store. I drove down the interstate at 70 MPH and had my wife sit in the passenger seat and operate the app. I have a 2014 Suburu Impreza hatchback and it measured 83 decibels. My first vehicle to test drive was a Hyundai Sonata. Quite a step up from the Impreza in terms of price, but I thought it would be worth it for a much quieter ride. Unfortunately the Sonata came in between 75 and 80 decibels which was quite surprising. The final car that I tested was a Nissan Rouge measuring 60 decibels. It was noticeably a quieter cab, but the app provided that reassurance.

Post a Comment

Your email address will not be published.

All comments posted become a part of the public domain, and users are responsible for their comments. This is a moderated site and your comments will be reviewed before they are posted. Read more about our comment policy »

TOP