Mailing List Follow akustyk on Twitter
Audio technology
Field recording
Film Processing

Review of the Audio-Technica BP892 omnidirectional headset microphone (for linguists)

July 2010


I would like to thank Audio-Technica U.S., Inc. for providing the BP892 microphone and offering valuable insight for the purposes of this review. In particular, I would like to thank Karen Emerson (Audio-Technica Media Manager) and Kurt Van Scoy (Audio-Technica Product Manager) for their help with the review. You can find information provided by Mr. Van Scoy quoted throughout the review.

Introduction: About the Audio-Technica microphone system

Established in 1962, Audio-Technica is one of the largest manufacturers of professional microphones, headphones, turntable cartridges, and other audio gear. Audio-Technica equipment has been used on the world’s greatest stages, in recording studios, government institutions, houses of worship, and many other venues requiring first-rate performance.

The BP892-TH MicroSet® Subminiature Omnidirectional Condenser Headworn Microphone (Figure 1) is a microphone that hooks behind the ear on either side of the face for a comfortable fit. This type of microphone is often referred to as an "earset" microphone, and several major manufacturers offer such microphones (e.g., the recently released Sennheiser Ear Set 1). They are designed primarily for unobtrusive and inconspicuous stage use. Such microphones are particularly suitable for live theater, music, and television production. The BP892 is essentially a series of microphones based on the same capsule, but offered in a few different terminations for both wired and wireless applications. The microphones are offered in two colors: black and beige.

Audio-Technica BP892 microphone

Figure 1. BP892-TH MicroSet® Subminiature Omnidirectional Condenser Headworn Microphone in beige

Over the last few years, the BP (Broadcast and Production) series of microphones emerged as a successor to the AT series of professional microphones. As the name suggests, the primary market for BP-series microphones is theater and live production where a broad dynamic range and improved resistance to high sound pressure levels (SPL) are a top priority. The BP-series microphones are engineered to have an improved resistance to high SPL (e.g., BP892's highest SPL value is 135 dB SPL) over their predecessors as well as competitor microphones, in an effort to provide Audio-Technica customers with a high-quality, reliable product. This feature is particularly important for microphones used in such close proximity to the sound source; the BP892 capsule being placed a mere inch or two from the talker's mouth

It is worth noting that the Audio-Technica BP896 lavalier microphone is based on the same capsule as the BP892, with a slight difference in sensitivity (about 1 dB) due to RF shielding. For all practical purposes the two microphones should have identical acoustic performance, which is great news for those of us who prefer not to use head-mounted microphones. It is quite understandable that some subjects may be reluctant to wear a microphone mounted around their ears during an interview. If you'd rather avoid putting your subject in an awkward position, you can simply use the BP896, clip it on to the subject's clothing (as close to their mouth as possible), and thus capture a recording of comparable quality, with just a very slight drop in signal-to-noise ratio due to farther placement.

The model reviewed here today the BP892-TH, which comes with the AT8539 11-52 V Remote Power Module and the Dual-Ear Microphone Mount (see Video 1). The kit is designed to be used with standard phantom power, available on most modern field recorders and microphone pre-amplifiers. Unlike Audix, Audio-Technica does not provide a battery-powered module off-the-shelf. However, on request, Audio-Technica will change the termination of the BP892's cable in order to make it compatible with the AT8537 battery-powered adapter (more on this in the section entitled "Battery power options" below).

Video 1. The unboxing of the Audio-Technica BP892 microphone

Design and specifications

The earpiece and the dual-ear mount

The BP892 is a lightweight, low-profile, earset microphone (Figure 2). It has a permanently attached wire headband and a plastic ear hook, which is shaped to fit comfortably behind either ear. The headband is an extremely important aspect of a field microphone. The subjects must feel comfortable putting the microphone on, the microphone must fit securely, and it must maintain a constant distance from the subject's mouth.

Audio-Technica BP892 microphone

Figure 2. The Audio-Technica BP892 microphone - earpiece, boom, and capsule

As has been often documented, some subjects feel self-conscious about recording their voices, so it is important to not escalate the feeling by using obtrusive instrumentation. To be perfectly honest, at first, I did not think the earpiece design would work, but after testing it for a few hours, I found it very easy to put on and, because of its extremely low weight and slim profile, I felt as though I were not wearing a microphone at all. So at least after the potential initial awkwardness, the subjects should feel perfectly comfortable with the microphone clipped behind their ear.

In Audio-Technica's in-house testing, this design was effective in about 75% of the cases. For all other cases, Audio-Technica designed the dual-ear headband as a modular option (Figures 3 and 4). The design was motivated by users' request for increased stability, the need to be able to fold the microphone flat, and for backwards compatibility with the AT series of microphones. Should you lose the extra earpiece, Audio-Technica can replace it through its service department.

Audio-Technica BP892 microphone

Figure 3. The Audio-Technica dual-ear headband

The dual-ear mount converts the earset design into the neck band similar to that of the Shure Beta 53 (review) and Beyerdynamic Opus 55 Mk II (review). After some initial reservations, I would rank the BP892's earset design very high in terms of comfort and ease of use. As far as I know, the design is unique to Audio-Technica. While I also do like the head mount of the Audix HT5 (review), the BP892 has the advantage of being able to fold the microphone and the dual-mount band flat and store it in a small box (Figure 5). The Audix HT5, while offering a superb fit, needs significantly more storage space.

Audio-Technica BP892 microphone

Figure 4.The Audio-Technica dual-ear headband hooks behind the ears and rests comfortably on the talker's neck

Audio-Technica BP892 microphone

Figure 5. The microphone, along with all the accessories, comes in a padded black box for easy storage

The microphone capsule

The BP892-TH is a subminiature omnidirectional condenser microphone with a broad and flat frequency response and a medium sensitivity (3.5 mV mV / Pa @ 1k). These features, alone, make it a good candidate for speech recording because it promises to capture clean and pure signals with rich spectral detail and a high signal-to-noise ratio. Detailed specifications can be found in Table 1.

Table 1. Audio-Technica BP892 Specifications
Element Fixed-charge back plate, permanently
polarized condenser
Polar pattern Omnidirectional
Frequency response 20-20,000 Hz
Low frequency roll-off 80 Hz, 18 dB/octave (wired only)
Open circuit sensitivity –49 dB (3.5 mV) re 1V at 1 Pa
Impedance 250 ohms (wired only)
Maximum input sound level 135 dB SPL, 1 kHz at 3% T.H.D.
Dynamic range (typical) 104 dB, 1 kHz at Max SPL (wired only)
Signal-to-noise ratio 63 dB, 1 kHz at 1 Pa
Phantom power requirements 11-52V DC, 2 mA typical (wired only)
Current consumption 0.1 mA typical at 5V (wireless only)
Voltage range 2.5-11V (wireless only)
Switch Flat, roll-off (wired only)
Weight Microphone, boom & earpiece:
2.6 g (0.09 oz)
Power module (wired only): 85 g (3.0 oz)
Dimensions Microphone: 8.1 mm (0.32") long,
2.6 mm (0.10") diameter
Boom: 98.4 mm (3.87") long,
1.07 mm (0.04") diameter
Power module (wired only): 97.6 mm
(3.84") long, 18.9 mm (0.74") diameter
Output connector (power module)
1.4 m (55") long (permanently attached
to microphone), 1.6 mm (0.06") diameter,
2-conductor shielded cable with locking
4-pin connector (wired only)
Audio-Technica case style M31
Accessories furnished
AT8539 power module (wired only);
AT8440 cable clip; AT8464 dual-ear
mount; two AT8157 windscreens; two
AT8156 element covers; moisture guard;
belt clip (wired only); carrying case

Audio-Technica BP892 frequency response

Figure 6. Audio-Technica BP892 microphone frequency response (according to the manufacturer)

The microphone capsule (Figure 7) is protected by a highly durable Parylene coating, it is built to stand up to the rigors of day-to-day use, including resistance to moisture, temperature, and handling noise. Audio-Technica made this particular design decision with aerobics, theater, and live stage performance in mind. Excessive amounts of sweat, especially when accumulated over a long period of repetitive use, may become a real hazard that may break down some of the electrical connections inside the microphone. "Audio-Technica recommends putting a bag of desiccant in the box or bag where the microphone is being stored, which will soak moisture out of the element and keep it more stable.” For most field recording environments, even the more extreme ones (e.g., a rain forest) there should be no problem with damage due to moisture.

Audio-Technica BP892 microphone

Figure 7. BP892-TH MicroSet® Subminiature Omnidirectional Condenser Headworn Microphone capsule

Modern professional condenser microphones are quite roadworthy. The BP892 is no exception, but Audio-Technica “does ask that people take care when bending the boom to avoid breaking it”. Also, “the protective cap should be used to keep make-up and other dirt out of the capsule.” Unlike some competitor designs (e.g., Sennheiser HSP2), the cap is for protection only. The cap “should be pressed all the way. Otherwise it would create an extra air space that may detune the microphone.”

The microphone cable is terminated with a locking 4-pin connector (Figure 8) for use either with Audio-Technica UniPak® body-pack transmitters or the AT8539 module. The locking connector is very well designed. It is easy to engage and disconnect. It provides additional security to the connection and helps prevent accidental signal interruption.

Audio-Technica BP892 microphone

Figure 8. The Audio-Technica locking connector used throughout the BP series

The AT8539 11-52 V Remote Power Module

The model reviewed here comes with the AT8539 11-52 V Remote Power Module (Figure 9), which enables the microphone to be powered by standard phantom power supplies. The range of operating voltage is rather broad (11 - 52 V), so the microphone should be adequately powered by most field recorders with built-in phantom power. In addition, the module has a very useful built-in high-pass filter (a.k.a, low-cut filter). The filter is well designed for speech applications because it only removes frequencies below 80 Hz and has a steep slope. The idea behind this type of design is to rid the signal being recorded of the low-frequency rumble that is ubiquitous in most industrialized urban environments. The filter removes frequencies below those critical for speech analysis, with the exception of male voices of low fundamental frequency. Unless your analysis relies on the frequency and amplitude of the first harmonic, the filter should not affect the accuracy of your spectral measurements.

Audio-Technica BP892 microphone

Figure 9. The AT8539 11-52 V Remote Power Module showing the low-cut filter switch (bottom image)

Figure 10 illustrates the influence of the low-cut filter on the inherent noise of a recording chain consisting of the BP892 microphone and the Sound Devices USBPre recording interface. As you can see, the peak at around 60 Hz is rather effectively removed by the filter, without too much change to the overall spectrum above 80 Hz or so. Even though I typically recommend disabling low-cut filters, this particular design (80 Hz, 18 dB/octave) might prove helpful for recording environments contaminated by the 50/60 Hz hum. Of course, one should try to identify all possible sources of noise before the recording session (see this article on noise), but if low-frequency noise cannot be otherwise avoided, the low-cut filter available with the BP892 microphone can save your precious audio data.

Comparison of two spectra of self-noise with and without the low-cut filter

Figure 10. A comparison of two spectra of BP892's inherent noise (with USBPre) without the low-cut filter (left panel), and with the low-cut filter enabled (right panel)

Battery power options

As I point out in every microphone review, it is rather desirable for a speech recording microphone to offer both standard phantom power and a battery power option. While we would most likely use stand-alone phantom power supplies in the lab, we would like to have the option to power the microphone with a AA or 9 V battery in the field, which helps save the portable recorder's precious battery life and provides the microphone with just the right amount of clean voltage for proper operation. The BP892 reviewed here comes with a phantom power adapter (AT8539), but if you look at the Audio-Technica catalog, you may not be able to find a matching battery powered unit. Fortunately, Audio-Technica does make the adapter (AT8537), except that it is designed for miniature condenser microphones terminated with a mini-XLR connector. I was assured by Mr. Van Scoy, Audio Audio-Technica Product Manager, that the BP892 can be re-terminated by the Audio-Technica Service Department to be compatible with the AT8537 adapter. This “would need to be a special order at the time the AT892cw/AT8537 were ordered.” The AT8537 performs the double duty of a battery power supply and a phantom power adapter, so if you do not want to use battery power, you can simply connect the unit to a standard phantom power source. I realize this can be a bit confusing, so I hope the diagram in Figure 11 may help explain it more clearly.

BP892 powering diagram

Figure 11. BP892 microphone's powering options

Audio tests

Low frequency response

Low frequency response is one of the critical features of a speech recording microphone. I must admit that I am somewhat biased in this regard because the bulk of my own acoustic analysis and synthesis lives below 500 Hz. However, if you are interested in the analysis of pitch, phonation, spectral tilt, F1, nasalization, breathiness, laryngalization, lateralization, and other articulatory features with low-frequency correlates, you will need a neutral low-end.

I tested the BP892 microphone with my usual setup. I generated a waveform with peaks of equal amplitude at the frequencies of 50, 100, 200, 300, 400, and 500 Hz. I then played the signal out of a flat-response loudspeaker and recorded it with the Audio-Technica BP892 microphone about one inch from the sound source. The spectrum in Figure 12 shows the low frequency response of the BP892 microphone. The microphone reproduces each center frequency really well, with no appreciable attenuation or amplification throughout the range. There is, however, a slight dip around 50 Hz, which is somewhat predicted by the manufacturer's specification, though perhaps not to such a degree. The dip is fairly insignificant and may be well within the possible analysis error. From my point of view, the Audio-Technica BP892 passes the low-frequency test with flying colors. If used with a flat-response microphone pre-amplifier, the BP892 will deliver accurate low-frequency response.

Low frequency response of the Audio-Technica BP892 microphone

Figure 12. The spectrum of low frequency response of the Audio-Technica BP892 microphone showing accurate response, with a slight dip around 50 Hz

Real-world self-noise performance

Since we record speech at relatively low sound pressure levels (50-60 dB SPL), we require the entire recording chain to be as quiet as possible (to have a low noise floor). Microphones of medium sensitivity (such as the BP892) may require your recorder's pre-amplifier gain to be turned up to perhaps 50-75% of its range, which, depending on the recorder, may or may not generate appreciable levels of self-noise. You can read more about testing equipment for self-noise in this post. The idea is to measure real-world inherent noise in the recording chain. I use the same methodology in all of my microphone reviews, so you can compare the present results with those obtained in the other tests.

The BP892 microphone turned out an outstanding self-noise performance and resistance to RF interference, on par with some of the major competitors, such as the Sennheiser HSP2 and Shure Beta 53. If you use a quality field recorder (e.g., Fostex FR-2LE or better), you should be able to capture recordings with a very low noise floor.

One word of caution: watch out for low-frequency hum. As you can see in Figure 13, there is a 60 Hz peak of around 50 dB. Low-frequency rumble is virtually unavoidable in any industrialized urban environment. In addition, your own equipment may generate 60 Hz (or 50 Hz outside of the US) hum due to ground loops or induction from power lines. This is especially true when using any of the popular USB recording interfaces with a laptop computer. The hum may come both form the interface and the laptop itself. Check your laptop's A/C adapter, as many of them are “dirty,” generating high levels of noise that leak into your precious recordings. You may want to try the Ebtech Hum X hum eliminating device. It does work wonders for hum originating from ground loops.

FFT of self-noise

Figure 13. FFT of the self-noise of the recording chain consisting of the BP892 microphone and the Sound Devices USBPre USB recording interface, calibrated to peak value of -12 dBFS and normalized to the RMS of 70 dB SPL

Resistance to high SPL

I tried overloading the microphone within normal-to-high conversational sound levels and prosody, including high-intensity plosives and laughter. I was not able to overload the microphone or cause signal distortion. With properly set recording levels, you can be confident that the BP892 will not overload easily. I do recommend, however, that you keep a windscreen on the capsule at all times (two are provided with the microphone) in order to avoid signal overload caused by rapid, high-energy airflow (e.g., plosives, wind, etc).

By the way, signal overload may also result from incorrectly set pre-amplifier gain. Try to calibrate your recorder (or its pre-amplifier, to be exact) to a peak value of between -18 and -12 dBFS so you do not overload the pre-amp. One of the most common mistakes in field recording is calibrating to dBFS values of -6 dB or higher. Not only do you cause more pre-amplifier noise, but you also run the risk of overloading it. If you have sufficiently high signal-to-noise ratio and record at a 24-bit bit-depth, you can calibrate much lower and increase the volume in post production, if need be. Simple volume increase is a linear process and will not unpredictably alter any spectral or temporal properties of your data. You can read more about setting recording levels in the field in this article.

Spectral detail

My audio tests revealed that the BP892 microphone is capable of perfectly good speech recordings both in the lab and in the field. The spectral detail is outstanding. You can discern very fine detail in spectrographic analysis both by eye and machine. Acoustic analysis with Akustyk 1.8 shows sharply defined formant peaks and a natural spectral envelope. In the temporal domain, there is enough detail to make even the most detailed analysis of, say, release burst, VOT, formant transitions, or phonation.

I put the samples through the re-synthesis algorithm available in Akustyk 1.8 to see if I can obtain a clean separation of source and filter data. Indeed, the samples produced excellent synthesis material. I would not hesitate to use the BP892 to record speech samples for the purposes of re-synthesis or for creating other types of auditory stimuli for the purposes of psychophysical experiments. Figure 14 shows formant tracks of the Polish word "bordo" before (black) and after re-synthesis (red). The F2 was increased by 100 Hz while F1 and F3 were decreased by 50 Hz throughout the utterance. Each analysis frame appears perfectly rendered, including the voiceless alveolar trill /r/, a very difficult sound to re-synthesize. You can download the samples at the bottom of this page.

Resynthesis example

Figure 14. Formant tracks obtained with re-synthesis (red) showing perfect rendition of each analysis frame

The spectrogram of the Polish phrase “czarna krowa” (Figure 15) shows excellent detail in both time and frequency domains. In particular, the low frequencies look really clean, and certainly promise accurate spectral analysis of nasalization or spectral tilt. Both oral and nasal peaks are very well-resolved and show excellent separation, making it possible to analyze their amplitude ratios with ease, which is the basis for nasalization analysis from narrow-band spectra (Akustyk 1.8).

You can see formant tracks in all the vowels as dark, narrow bands of energy, which the BP892 captured with superb accuracy. The same level of accuracy is evident in the high-frequency release burst spectra, which show no trace of signal overloal, popping, or clipping.


Spectrogram of a recording obtained with the Audio-Technica BP892 microphone

Figure 15. A spectrogram of the Polish phrase "czarna krowa" obtained with the Audio-Technica BP892 microphone and the Sound Devices USBPre at 48,000 Hz, 24-bit, calibrated to peak level of -12 dBFS

The FFT of the vowel /o/ in “krowa” (Figure 16) shows a very natural frequency envelope as does the LPC graph in Figure 17. The formant peaks are very well defined and promise accurate formant analysis.


Figure 16. FFT of the vowel /o/ in the word "krowa" showing sharply defined formant peaks and a natura envelope


Figure 17. LPC of the vowel /o/ in the word "kropki" showing very well-defined formant peaks and narrow formant bandwidths. The amplitude of F4 seems a bit high, but this is strictly due to the the idiosyncrasy of the talker's spectral tilt rather than a property of the microphone itself.

The phrase "buy a large barrel of good beer" in Figure 18 was recorded with the Audio-Technica BP892 microphone, a state-of-the-art Tucker-Davis microphone pre-amplifier, and the Sound Devices USBPre at 24-bit/48,000 Hz in a sound-proof booth. The recording shows exemplary spectral detail and a very pleasing, crisp subjective sound quality. You can download a recording of thirty test sentences in the section "Download audio files" at the bottom of this page.

Listen to MP3 at 128 kbps:

Spectrogram of a phrase recorded with the BP892 microphone and Tucker-Davis pre-amp

Figure 18. A spectrogram of the phrase "buy a large barrel of good beer" recorded with a state-of-the-art Tucker-Davis microphone pre-amplifier

Finally, let's look at a recording made with the BP892 microphone and the Fostex FR-2LE field recorder. On paper, the microphone and the recorder appear to be properly matched (see this tutorial for more details). In my tests, I found this to be true. The Fostex FR-2LE provides a clean, quiet, and stable voltage to the microphone. As a result, the microphone/recorder chain produces high-quality 24-bit PCM files. I found both the detail and the subjective sound quality to be perfectly satisfactory. Figure 19 shows a spectrogram and a waveform of a Polish phrase "mama ma mate." The phrase contains mostly nasalized consonants and vowels, which, to be properly rendered, require a high-precision microphone with a neutral frequency response below 500 Hz. The Audio-Technica BP892 produced an audio file with very rich and accurate detail, including an excellent separation of oral and nasal peaks showing smooth transition of nasality from the consonants to the vowels. You can download the audio file in the "Download audio files" section below.

Spectrogram of a Polish phrase "mama ma mate" with Audio-Technica BP892 microphone and Fostex FR-2LE recorder

Figure 19. A spectrogram of a Polish phrase "mama ma mate" with the Audio-Technica BP892 microphone and the Fostex FR-2LE recorder showing excellent low-frequency detail


Having tested the Audio-Technica BP892 microphone thoroughly in the lab and in the field, I have come to really appreciate its superb performance and unique design. The microphone proves to belong in the elite league of speech recording microphones. I personally found the microphone to be somewhat similar to the Sennheiser HSP2 and Shure Beta 53 in terms of design and performance. In purely subjective terms, there is a certain crispness to the way speech recorded with the BP892 sounds. It is both highly intelligible and pleasant to listen to. The microphone would make a fantastic tool for making speech corpora for the purposes of language documentation and long-term preservation. Such corpora would also make superb testing material for speech engineers and speech stimuli for hearing scientists. I can also definitely see the Audio-Technica BP892 used as a diagnostic tool for speech and language pathologists in a clinical setting. Finally, oral historians, for whom the subjective sound quality is of paramount importance, should give this microphone a serious look, either the earset (BP892) or the lavalier version (BP896).

Being part of the elite is an excellent achievement in and of itself, but the BP892 has important competitive advantages over the likes of Shure Beta 53 or Sennheiser HSP2. First, field recordists will be very happy to know that they can use the same capsule either as a lavalier (BP896) or a head-mounted microphone (BP892). This type of versatility will be of particular benefit with subjects who may be reluctant to wear a head-mounted microphone and would prefer the less obtrusive placement of a lavalier. Second, the BP892 can be used with a battery-powered adapter (AT8537), but you need to make sure to request that the BP892 be re-terminated to fit the adapter when you order it. The AT8537 adapter has a somewhat more attractive and less bulky design than the more common belt-pack and it powers the microphone with a single AA battery. Third, the BP892 has a very interesting, unique earpiece design, with the optional dual-ear mount and the ability to fold down flat for easy, space-saving storage. Finally, Audio-Technica has an exceptionally good worldwide sales presence and technical support.

I have been using Audio-Technica microphones since 1998 and I have found both the products and the customer service to be exceptionally good. Audio-Technica products are competitively priced and are positioned to fill many different needs with unique designs and specifications. For instance, the Audio-Technica ATR3350 is a great lavalier microphone of unmatched quality in its price range. The AT831b is one of the best oral history microphones available today, with a directional polar pattern, a battery belt-pack, and a low-cut filter. Similarly, the AT8022 is one of the best, inexpensive stereo field recording microphones I have ever used, compatible with most portable digital recorders and camcorders with a 1/8-inch stereo microphone input. The list goes on.

As far as acoustic performance is concerned, most of the omnidirectional head-mounted condenser microphones reviewed on this website are capable of making very good speech recordings. Your purchasing decision should be motivated by the features that fit your individual needs and by the quality of customer support offered in your country. For me, personally, the BP892 has a lot of appeal. I particularly like the fact that I can use a lavalier and an earset microphone interchangeably, and obtain nearly identical acoustic performance. However, as always, the decision is yours.

Download audio files

  Description Quality Comments
Download With Tucker-Davis pre-amplifier and Audio-Technica CP8506 48VDC 4-Channel Phantom Power Supply 48,000 Hz; 24-bit; mono Captured with USBPre in a sound booth
Download With the USBPre; test recording in Polish 48,000 Hz; 24-bit; mono Captured with USBPre in a sound booth
Download With Fostex FR-2LE and onboard phantom power 48,000 Hz; 24-bit; mono Captured with Fostex FR-2LE, calibrated to peak level of -12 dBFS, in a quiet room