The basic equipment you need to record sound are a microphone, some electronics to amplify and digitise the signal and a means of storing the recording. It’s also a good idea to monitor what you are recording for which you will need headphones. A smartphone or voice recorder will incorporate all of these but they are not usually good enough to pick up the quieter sounds of wildlife in the field. Pretty soon you will want to construct your own system from the constituent parts. Most of the equipment on the market was designed for the music or film industries, whose needs are often a little different from ours, so we need to pick and choose carefully. Below we run through a recordist’s basic equipment to help you with those choices.
A microphone is any device that converts sound waves into electrical signals. There are many ways of achieving that. You’ll read about dynamic, ribbon, carbon, piezoelectric and countless other types but much the most commonly used and practical for wildlife recording are condenser microphones. We’ll only deal with those here. A condenser (or capacitor) is a simple electrical component consisting of two sheets of conducting material separated by insulating material and holding an electrical charge between them. The electrical signal a condenser microphone produces results from the change in voltage between the two sheets that occurs when sound waves distort them.
The means by which the voltage across the condenser is set up is the basis of the difference between two main types of condenser microphone. In the conventional type the voltage is simply generated by a power supply. In the second type, the electret microphone, the voltage is electrostatic, set up and “baked in” when the device is manufactured. You’d think the charge would soon leak away but, in fact, it is so effectively sealed within insulating material that it is good for hundreds of years. Electret microphones are small, robust and cheap, and modern ones are remarkably sensitive and with low self-noise – perfect for a beginner dipping in a toe without committing to great expense. They do not, however, have the dynamic range or the capacity to capture the detail that a conventional condenser microphone has. Many factors determine the quality of a microphone including its robustness for field use, the diaphragm size (bigger more detailed) and impedance (lower the better); in general you get what you pay for.
One particular curse of the condenser microphone is its susceptibility to humidity and condensation. This does not necessarily destroy the mic but will incapacitate it until it is dried out. A third variant, the RF condenser microphone, has been introduced to reduce this problem. While these also employ a condenser to detect sound waves it is utilised in a different way: they set up a rapidly oscillating electrical circuit whose frequency is determined by the size of the gap between the condenser sheets. Thus sound modulates the frequency rather than the amplitude of the voltage, which makes them far more robust for use in the field. Further circuitry is required to generate the oscillation and convert the frequency change to a voltage change that the recorder can understand, all of which makes these microphones quite a lot more expensive.
Power and cables. Most microphones have a built-in amplifier circuit required to convert their tiny voltage changes into something strong enough to be reliably passed
In most cases the signal travelling between the microphone and the recorder is weak and will require considerable boosting by the recorder’s pre-amp. The cables themselves can be long and could act as an aerial picking up any electrical interference from mains hum to Radio Moscow, which will, of course, also be amplified by the pre-amp. To prevent this problem XLR cables are balanced. This involves not just shielding and twisting of the wires to minimise interference, but also a cunning piece of circuitry that subtracts the noise. Two copies of the signal are passed down separate wires, in one of which the voltage has been reversed. As both signals pass along the cable both suffer the same interference. Then, at the receiving end, the reserved copy is flipped back to its original polarity. Now the signal in both wires is of the same polarity while the polarity of the noise is reversed so combining them neatly cancels out the noise. Unbalanced cable of the sort often used to provide PiP is not so protected.
Frequency response. We are lucky because, unless you intend to record bats or elephants, most wildlife uses the same range of frequencies as the human ear detects. Actually, this is probably no coincidence because the human ear is adapted to hear most of what is going on in the natural world. And, in any case, the sounds of the natural world we find interesting are obviously those we can hear: ultra- and infra-sound, scientifically fascinating though they may be, do not form part of those majestic dawn choruses, or spine-tingling calls echoing through the forest, that we find so attractive. So most microphones will pick up an appropriate range of frequencies. We also share with most other microphone users the desire for a flat response that doesn’t emphasise some frequencies over others. So, in this respect, what we want is usually available commercially. But again, you get what you pay for: more expensive microphones tend to have a flatter response over a broader range of frequencies.
Microphone directionality. Not all microphones pick up sound equally from all directions. Those that do are known as “omnidirectional”, or “omnis”. Other common types are cardioid (essentially one-sided), super- and hyper-cardioid (or shotgun – reject sound from all but a narrow angle in the direction that the mic is pointed), bidirectional (or figure-of-eight – picks up sound on either side of the plane of the microphone but rejects sound in its plane).
Microphones that pick up sound from a particular direction obviously allow you to record a particular subject while rejecting noise from elsewhere. Long, shotgun microphones are the best at this. They are not, however, a panacea for the problem of environmental noise. Noise will reach the microphone both because the microphone’s off-axis rejection is only partial and because noise is reflected off objects in the environment.
Directional microphones do not amplify sounds from the direction in which they are pointed; they only reject sounds from other directions. A parabolic reflector with a microphone set at its focal point does amplify sound from a narrow direction. The larger the dish the greater will be the amplification, though clearly there is a trade-off with practicality (about 56cm is found to be a good compromise). The angle within which the signal is amplified is narrow so it is a good idea to use a transparent dish to avoid obscuring your view of the subject. A parabolic dish offers an effective and simple means of isolating a songbird’s song and can be used to make excellent recordings. The beginner will appreciate the instant success this brings. Dishes do, however, “colour” the sound: they amplify higher frequencies much more than lower ones and any sound with a wavelength below that of the width of the dish will not be amplified at all. This problem is even more pronounced for off-axis sounds, which can sound quite distorted and strange. Commercially available dishes can be a little expensive for the beginner but DIY solutions are possible.
Stereo. Isolating a single singing bird’s song is not everyone’s objective. Stereo recording can provide a far fuller, more immersive sound picture of a whole habitat. They are also capable of separating out different sounds that in a mono recording may blur together. For this reason noise can seem less intrusive, although generally harder to avoid, than in a mono recording.
The brain uses two clues to work out the direction a particular sound is coming from: phase and amplitude. A sound coming from your left will arrive slightly earlier and sounds louder in your left ear. There are numerous stereo techniques but all utilise timing and/or intensity differences. Here are some commonly used techniques:
Spaced pair (or A/B configuration): two omni or cardioid microphones set 1-3 meters apart.
Coincident pair (or X/Y configuration): two cardioid microphones set at an angle (often 90°) to one another and as close together as possible.
Mid-side (M/S) arrangement: a cardioid microphone facing the subject and a figure-of-eight (or bidirectional) microphone at right angles and as close as possible to this. The use of the fig-of-8 mic and need to decode the signals seem awkward but are repaid by allowing you to adjust the balance between the main subject and the surrounding ambience. The Blumlein configuration is similar but involves two figure-of-eight microphones.
ORTF (Office de Radiodiffusion Télévision Française) technique: two cardioid microphones separated by 17cm and facing at an angle of 110° to on another.
Binaural dummy head: a dummy head with anatomically correct (silicone) ears within each of which is place a small omni microphone.
All of these can produce effective stereo sound images but each has its pros and cons. A problem common to all techniques that involve microphones that are spaced out is that, when combined as a mono track, they suffer interference: some wavelengths cancel out while others are reinforced. (Similar problem occurs when the listener is not equidistant from stereo speakers.) Those techniques that use coincident microphones (X/Y and M/S) do not suffer this, although they do lose all timing information. The rather bizarre binaural head is in some ways the most realistic, especially when recordings are played back through headphones. It nevertheless is limited by the size of the microphones that can be fitted in the ears of the dummy.
Wind. The slightest breeze will cause an awful roar on any unprotected microphone. Many microphones come with an open-pore foam windshield that will deal with light winds. But more wind than that requires the mic to be enclosed in a cage surrounded with open-weave, furry material. These “blimps” can be astonishingly expensive to buy. Fortunately, equally effective, homemade alternatives can be constructed for a fraction of the cost. A wire birdfeeder, covered with a stocking or faux fur and the microphone secured in place with elastic bands works well. The important point is that the covering should be acoustically transparent across all wavelengths.
A vast and ever-changing range of portable recording devices is available. Given that the recorder is probably the most important and expensive piece of kit the recordist will own, this can be quite daunting. Before deciding the most appropriate for you it is necessary to understand the many functions of the recorder.
Input from microphone. All but the very cheapest recorders include sockets to attach external microphones. These, as we have seen above, come in two sorts: stereo jack sockets for unbalanced cable and XLR for balance cable. The cheaper recorders will only provide the former with plug-in-power (PiP) while more expensive devices will also provide two or more three-pin XLR sockets with phantom power. Many also have third input socket, “line in”, which expects an already-amplified input and skips the device’s pre-amp.
Preamplifier. For the quiet sounds of wildlife this is often where the game is won or lost. At the cheaper end of the market, preamp noise is often a much greater problem than that from the microphone. The further you turn up the pre-amp gain, the noisier it becomes. Thus even a microphone with a good signal-to-noise ratio, if it is quiet, can be unusable due to the noise introduced by the preamp. Fortunately, recorders with very clean pre-amps are becoming increasingly affordable. This area is changing rapidly so seek advice on what are currently the best.
Digitising. Once suitably amplified the signal is digitised: it is sampled thousands of times per second and its value at each sample recorded as a digital number. It is now essentially immune to degradation or acquiring further noise. The frequency of sampling (the “sampling rate”) and the precision with which each sample’s value is recorded (the “bit depth”) determine the quality of the recording. Common settings are those used for audio CDs: 44,100 samples per second (44.1 kHz) and 16 bits per sample (allowing 65,536 intensity levels). Some recorders will allow greater sampling rates and bit depths for improved definition but, of course, these produce larger data files. Fiddling with these values is something for the experts and need not bother the beginner.
Storage. Almost all digital recorders these days store the digitised signal on standard SD (Secure Digital) cards like those used in cameras and other digital devices. If you’d like to make lengthy recordings you may want to replace the SD card that comes with the recorder with something larger, but before buying a whopping one check that the record can read it. At the standard sampling and bit depth 1 hour’s stereo recording will occupy approximately 0.6 GB. To save space most recorders will allow you to store the recording in compressed (usually MP3) format. While this inevitably results in some loss of information, nevertheless huge space savings can be achieved with no discernable loss of quality.
Controls and display. Even the simplest of recorders will have innumerable settings. All have some sort of menu system to control such things as the sampling rate and bit depth, which input channels to record from and whether to record a stereo or mono signal, whether to compress the data, whether to automatically control the gain (rarely advisable for our purposes!), to switch on “pre record” (i.e. continually record a buffer of a few seconds to capture sounds immediately before the record button is pressed) and setting which type of batteries are being used (this mostly so that the battery life display is accurate) and the time and date (don’t forget this whenever replacing the batteries). Each device has its own system of buttons and knobs to navigate around the menu, which can be a little fiddly and counterintuitive if you aren’t used to it. Matters are often made worse by the quality of the display; it is worth checking that this can be read outdoors in daylight. More sopisticated machines have more options so, to make life easier, often provide a number of “presets” which, once set up, allow a whole suite of setting to be selected at once. There will often also be several knobs that allow you to adjust the preamp gain and the headphone volume. Finally there will be three standard buttons to start recording, stop and replay. All in all, recorders can be quite complicated to use so it is essential that you spend some time getting to know all its way before taking it out into the field.
Batteries and battery life. Almost all portable recorders run on AA batteries (one or two use a Li-ion battery). These may either be disposable or rechargeable. Modern nickel metal hydride (Ni MH) rechargeable batteries store so much energy (up to 2800mAh) and hold their charge so well that it is hard to see why anyone would ever bother with single-charge batteries. However, be warned that they have a small fire risk and rechargeable batteries should never be carried in hold luggage on an aeroplane. Recorders themselves vary enormously in the power they require: some will record for 24 hours without draining the batteries while others, especially when phantom power is used, barely last a couple of hours. This means obviously that some recorders simply are not suitable for lengthy, unattended recording. And it is always worth carrying plenty of fully charged spares. Many recorders can be powered from additional, external sources, perhaps the most practical of which for field recording being Li-ion power packs usually used to charge mobile phones.
Once you have your recording safely digitised and on your computer you will want to polish it up a little. It may need amplifying again and noisy or uninteresting bits trimmed off. You might also attempt a little filtering (perhaps taking out frequencies below 100 Hz) to reduce wind or traffic noise. All of this can easily be done with free software available on the Internet. More ambitious cleaning or mastering of the recording is something of a black art and requires more sophisticated and expensive software. There are different schools of thought about how far down that road wildlife sound recordists should go. For some interfering with the “natural” recording is an anathema; for others what the microphone captures and what the ear hears are two different things and returning the former to the latter is an essential objective. Certainly over-use of filtering or noise reduction can simply sound weird. But, done well, who wouldn’t prefer a wildlife recording with that yapping dog or passing motorcycle removed? And what harm is there in sometimes enhancing natural sounds to make them more dramatic or emphatic? In the end it’s up to you and how scientific or artistic you want to be.
In all probability your first recordings will be disappointing, spoilt by hisses and roars, and bumps and rustles. You need to learn the when, where and how to record; to take account of conditions and the habits of the creature you hope to record. How far away and how busy are the nearest roads? Each tree leaf or reed spear rustles quietly but in a wood or reed bed with more than a breath of wind millions combine to produce a din that may overwhelm your recording. A good recording requires planning and ears that hear and assess all the sounds in the environment.
Plan as you might, environmental noise is almost impossible to avoid completely. It can be mitigated to some extent with a shotgun microphone or parabolic reflector, but these, as we’ve seen, have limitations. Better to place your microphone as close to the subject as possible without disturbing it. That’s surprisingly easy to do with some small songbirds but most require a little cunning and patience. Observe your subjects and get to know their habits. Then, on a calm day before the traffic gets going, set up your microphone where you know they will return, withdraw, hide and watch and wait with your recorder on the end of a long cable. Or, simply leave your kit recording unattended. Although it can be frustrating afterwards to work out who was doing what, this is an excellent method to capture the natural soundscape of a habitat. You’ll be amazed what goes on when you aren’t there! Obviously your presence can be inhibiting but what about the presence of your microphone placed so close to your subject? A furry wind muff can look a lot like a predatory mammal, so toss a piece of camouflaged scrim over it.
Sometimes there is no alternative to pointing a handheld microphone at a bird encountered by chance. That’s when you discover how noisy you are: all that breathing, coughing and stomach rumbling. Your bones creak – everyone’s do. It helps to cradle your mic in a “pistol grip” to isolate it from handling noise, but you still need to be ultra still and careful. What about that cable swinging about between the mic and the recorder? The sound of everything it touches is transmitted directly to the mic. For a scientific study you may want quantity rather than perfection and handheld microphones are the only practical method. A few yards of cable between you and the microphone make a huge difference but it does curb spontaneity.
There are many more lessons to learn. A large part of the fun of sound recording is in overcoming the challenges posed when trying to capture natural sounds. When you’ve come up with a new solution of your own, you can head over to the WSRS Forum and tell other recordists about it.