Why audio vocabulary matters
Sound has its own vocabulary, and much of it describes things you cannot see. A recording can sound thin, boomy or harsh, and the words below give you a way to talk about why. These audio terms are grouped by where they come up: recording, editing and exporting. The definitions are simplified, because the exact numbers and behavior of equipment vary.
Recording terms
Level. How strong an audio signal is. When people say “check your levels,” they mean look at how loud the signal is as it enters the recorder. Levels are shown on a meter, usually in decibels.
Decibel (dB). A unit for comparing sound or signal strength. In digital audio, the scale tops out at 0 dB, and values below it are written as negative numbers, such as -12 dB. The scale is not linear, so a small change in decibels is not a small change in perceived loudness.
Gain. The amount of amplification applied to a microphone’s signal before it is recorded. Too little gain leaves the recording quiet and noisy once you turn it up later; too much causes distortion. It is not the same as playback volume.
Clipping. Harsh distortion that happens when a signal is too strong for the system to record, so the peaks of the sound are flattened off. Clipping is usually not repairable well, so it is better to avoid it by leaving some headroom.
Headroom. The space between your usual level and the maximum a system can handle. Leaving headroom means a sudden loud laugh will not clip.
Noise floor. The constant background sound present in a recording even when nobody is speaking, such as electrical hiss, a computer fan or air conditioning. Anything quieter than the noise floor is buried in it. A lower noise floor generally means a cleaner recording.
Signal-to-noise ratio. A comparison of the sound you want against the noise floor. A higher ratio generally means the voice stands out more clearly.
Room tone. The natural sound of a quiet room with no one speaking. Editors often record a short sample so they can fill gaps smoothly.
Plosive. A burst of air from sounds like “p” and “b” that makes a thump in a microphone. A pop filter or better positioning reduces it, which is part of general microphone technique.
Monitoring. Listening to the audio as you record, usually with headphones, so you can catch problems immediately.
Digital audio terms
Sample rate. How many times per second a digital recording measures the sound wave, written in hertz or kilohertz. Common values include 44.1 kHz and 48 kHz. A higher rate captures higher frequencies, but for speech the difference is often hard to hear.
Bit depth. The number of values available to describe each measurement of the wave. A greater bit depth gives a larger range between the quietest and loudest sounds a recording can hold, and generally more room for editing.
Bitrate. The amount of data used per second in a compressed audio file, usually in kilobits per second (kbps). A lower bitrate makes a smaller file, but may reduce quality. Bitrate describes the file’s data rate; it is not the same as sample rate or bit depth.
Codec. The method used to compress and decompress audio. Some codecs discard information to save space; others do not.
Lossy and lossless. Lossy compression permanently throws away some audio data to make smaller files. Lossless compression shrinks the file without discarding anything, so the original can be reconstructed exactly.
Mono and stereo. A mono recording has one channel. Stereo has two, usually left and right. Speech is often recorded and published in mono, though stereo is common for music.
Editing and mixing terms
Normalization. Adjusting a whole recording by a fixed amount so its peaks or average loudness reach a chosen target. It does not change how dynamic the audio is.
Compression (dynamic range). Do not confuse this with file compression. Dynamic range compression reduces the difference between the loudest and quietest parts, so speech sounds more even. Used heavily, it can make audio sound flat.
Equalization (EQ). Boosting or cutting particular frequency ranges to change the tone, such as reducing rumble or adding clarity.
Noise reduction. Software that lowers steady background noise. It works best when the noise is constant, and strong settings can make a voice sound watery or artificial.
Fade. A gradual rise or fall in volume, used at the start or end of a clip.
Crossfade. A transition in which one clip fades out while the next fades in. It is a standard tool in audio editing.
Ducking. Automatically lowering background music while someone speaks, then raising it again.
Mix. The balance of all the sound elements, including voices, music and effects, in a finished piece. Careful sound design makes the mix feel intentional.
Loudness. How loud audio seems to a listener over time. It is measured differently from peak level, and some platforms adjust playback to a target loudness. Targets differ, so check the current requirements of wherever you publish.
Exporting terms
Export or render. Producing a finished file from an editing project.
Metadata tags. Information stored in an audio file, such as title, artist and episode number.
Transcript. A text version of the spoken words. Providing audio transcripts helps people who cannot listen and makes content easier to search.
Putting the terms to use
If you are setting up a quiet space for the first time, home recording usually comes down to three ideas from this list: keep the noise floor low, set gain so your levels are healthy, and leave headroom. Those choices shape everything you do later during audio editing. Most other terms matter more once you are polishing a finished piece.
Results vary with equipment and rooms, so treat any numbers you read as starting points and let your ears confirm them.