In addition to some of the basic design characteristics we've discussed, there are some technical specifications which are particularly associated with microphones - let's look at the most important to understand:
Numbers alone will not tell you that much in this area, as not many microphones have such a limited frequency response that they do not cover the area corresponding to human speech! More useful is a frequency response graph which displays the amplitude of response at any given frequency, generally only for sound directly into the front of the microphone, although sometimes additional graphs are given for 'off axis' response. Although you might expect an ideal microphone to have a straight-line graph, it's variation in frequency response that is mainly responsible for giving different mics their distinctive characters. Microphones aimed at speech use will often have a lift in the midrange to aid intelligibility - look for a boost between around 500Hz and 3kHz - not much higher though or excessive sibilance (over-emphasised 'S' sounds) will ensue.
Sensitivity expresses the microphone's ability to convert the movement of its diaphragm to electrical voltage. To accurately measure the sensitivity of a microphone, a manufacturer will place it in a 'reference sound field' to measure the output voltage against a known sound pressure level (see below), and this stated as its sensitivity, usually in mV/Pa where higher numbers are better, or sometimes in dBV, where confusingly figures are negative, and numbers closer to zero (i.e. less negative) are better. In a situation where you're recording low-level signals as is common with speech, particularly where the microphone will be a distance away from the speaker, it's important to use a mic with high sensitivity (and therefore high output) in order to keep noise levels relatively low in any following circuitry.
A microphone, like all electronic devices, generates noise. The amount though is usually well below the noise level of other equipment, and so generally isn't a factor when recording loud or moderately loud sound sources. However, microphone noise can be a problem when recording very quiet sound sources. Manufacturers are aware of this problem, and many mics are now advertised as having low noise. It is usually termed 'self noise', 'equivalent noise SPL' or 'noise floor' in specifications, and is the audible noise level the microphone produces when it's placed in isolation from external sound sources. Dynamic mics usually have particularly low self-noise as they don't have the electronic component count of condenser mics.
Sound Pressure Level (SPL)
Sound Pressure Level is a reference standard that indicates how loud a sound is, measured in decibels or 'dB'. The ability of a microphone to handle high SPLs without distortion is primarily determined by its basic type. Dynamic mics are generally the most robust in this respect, and can usually handle SPLs of 140dB or more. Ribbon microphones are the weakest and can actually be destroyed by too high a level. Although the diaphragm of most condenser mics generally won't distort except under the most severe SPLs, their built-in pre-amps can be overloaded to the point of distortion, so figures vary widely depending on their quality. Many condensers have an attenuation 'pad' switch though, which reduces the level of the signal before the pre-amp to avoid overloading. When using the pad, keep in mind that the signal-to-noise ratio is degraded by the amount of attenuation, so in normal SPL conditions it's wise to switch out the pad - this shouldn't be needed in most reporting situations in any case, unless you're in a war zone!