Microphone

• The collection of extraneous noise. This can be a concern, especially in amplified performances, where audio feedback can be a significant problem. Alternatively, it can be a desired outcome, in situations where ambient noise is useful (hall reverberation, audience reaction.)

• Choice of a signal type: Mono, stereo or multi-channel.

• Type of sound-source: Acoustic instruments produce a very different sound than electric instruments, which are again different from the human voice.

• Processing: If the signal is destined to be heavily processed, or "mixed down", a different type of input may be required.

• In close miking, a directional microphone is placed relatively close to an instrument or sound-source. This serves to eliminate extraneous noise-- including room reverberation-- and is commonly used when attempting to record a number of separate instruments while keeping the signals separate, or when in order to avoid feedback in an amplified performance.

• In ambient or distant miking, a sensitive microphone or microphone is placed at some distance from the sound source. The goal of this technique is to get a broader, natural mix of the sound source or sources, along with reverberation from the room or hall.

• The X-Y technique involves the coincident placement of two directional microphones. When two directional microphones are placed coincidentally, typically at a 90+ degree angle to each other (typically with each microphone pointing to a side of the sound-stage), a stereo effect is achieved simply through intensity differences of the sound entering each microphone. Due to the lack of time-of-arrival stereo information, the stereo effect in X-Y recordings is somewhat unnatural, especially when listened to through headphones. The main advantage is that the signal is mono-compatible, i.e., the signal is suitable for playback on non-stereo devices such as radio.

• The Mid-Side (M-S) technique is a special case of X-Y and uses a directional or omnidirectional microphone (M) and a bidirectional (figure-8) microphone (S), placed at a 90 degree angle to each other with the directional microphone facing the sound-stage. The outputs of these microphones are mixed in such a way as to generate sum and difference signals between the outputs. The S signal is added to the M for one channel, and is subtracted (by reversing phase and adding) to generate the other channel. M-S has two advantages: when the stereo signal is combined into mono, the signal from the S microphone cancels out entirely, leaving only the mono recording from the directional M microphone; additionally, M-S recordings can be "remixed" after recording to alter or even remove the stereo spread. The M-S technique with an omnidirectional M microphone is equivalent to X-Y with two cardioids at a 180-degree angle.

• Near-coincident recording is a variant of the X-Y technique and incorporates inter-aural time delay by placing the microphones several inches apart. The ORTF stereo technique calls for a pair of cardioid microphones placed 17 cm (7 inches, which is approximately the distance between the ears) apart at an angle of 110 degrees. In the NOS stereo technique, the angle is 90 degrees and the distance is 30 cm. The choice for one of the other dependens on the distance to and the width of the sound source that is recorded; ORTF is better for short distances. This technique leads to a realistic stereo effect, for playback through both headphones and loudspeakers. It has a reasonable mono-compatibility.

• The A-B technique uses two omnidirectional microphones at a large distance (several meters). Stereo information consists in large time-of-arrival distances and intensity differences. On playback, the stereo image is usually perceived as somewhat unnatural, as if the left and right channel are independent sound sources, without an even spread from left to right. A-B recordings are unsuitable for mono playback because the time-of-arrival differences can lead to certain frequency components being canceled out and other being amplified, the so-called comb-filtering effect.

• Binaural recording is a highly specific attempt to recreate the conditions of human hearing, reproducing the full three-dimensional sound-field. Most binaural recordings use model of a human head, with microphones placed where the ear canal would be. A sound source is then recorded with all of the stereo and spatial cues produced by the head and human pinnae. Binaural recording is usually only somewhat successful, in addition to being highly inconvenient. For one thing, it tends to work well only when played back directly into the ear canal, via headphones (no speakers), as other methods of playback add additional spatial cues. Furthermore, as all heads and pinnae are different, a recording from one "pair of ears" will not always sound correct to another person. Also, headphones have a frequency response that compensates for the fact that the reflections from the pinnae, head and shoulders strongly affect the frequency spectrum, with the assumption that a recording is taken with a flat frequency spectrum. Introducing the spectral distortion already in the binaural recording results in an unnatural frequency spectrum, even when played through headphones. Finally, as visual cues are generally much more powerful than auditory cues when determining the source of a sound, binaural recordings are not always convincing to listeners.

• Microphones: How They Work
• Microphone pictures and specifications at coutant.org.