Glossary Terms

180-degree phase shift

See phase inversion

4-pole filter

A class of electronic or digital filters whose frequency response exhibits a maximum change of 24dB per octave (before taking into account resonance). This is the most common type of filter used in analog music synthesizers.

See also filter.

4/4 time

This time signature indicates that there are four quarter note (crotchet) beats per bar.

See also time signature.

AC

Abbreviation for Alternating Current. In a sound synthesis context this term usually refers to an oscillating signal representing audio, as opposed to a slow moving control signal such as an envelope, which might be considered DC or Direct Current.

accent

Emphasis given to a musical note. An accented note is usually louder than those surrounding it. An accent can be used to emphasize the important notes in a particular metre/time signature (each metre has its own prescribed rules about emphasizing particular beats), but can also be used to accent other notes.

additive synthesis

A sound synthesis method where spectrally rich timbres are created by mixing together a number of simple sounds, usually sine tones. It is called "additive" in contrast to "subtractive synthesis" where a spectrally rich sound is filtered to shape its timbre.

ADSR envelope

ADSR stands for Attack-Decay-Sustain-Release. It refers to an envelope shape which is commonly used in sound synthesis to control amplitude, pitch or filtering over time. It is a crude approximation of the time dynamics of natural sounds. Each of A, D, S, and R is called a segment. In an ADSR envelope, the user controls the duration of the Attack, Decay and Release segments and the level of the Sustain segment. Usually the Sustain segment lasts as long as the musical note is sustained (for example, by keeping a key depressed on a keyboard).

See also modulation.

all-pass filter

A type of filter which passes all frequencies, often used as a building block for other types of filters. All-pass filters typically delay different frequencies by different amounts, causing frequency-dependent phase changes. As a result, mixing an all-pass filtered signal with its (unfiltered) input will usually lead to phase cancellations and reinforcements at different frequencies -- thus creating a filter which exhibits a frequency dependent amplitude response.

amplitude

A measure of the "strength" of a digital or electrical representation of an (audio) signal. Corresponds to the height of an audio waveform. It roughly corresponds to the perceptual quantity of loudness. Amplitude may be measured in volts, but more commonly in decibels (dB), a logarithmic measure relative to an absolute quantity such as a fixed voltage (dBV) or the maximum representable amplitude in a digital audio system (dBFS).

amplitude gating

see gating.

amplitude modulation

The process of changing (modulating) the amplitude of a signal with another signal. The process has certain well understood spectral properties -- for example, the resultant signal will contain frequencies at the sum and difference of frequencies present in the inputs.

See also modulation.

analog

A process or sound synthesis method implemented using analog electronic hardware as opposed to digital methods implemented using digital circuitry or computer software.

arpeggio

A sequence of musical notes corresponding to the notes in a chord, but played one after another instead of all at once.

arpeggiator

AudioMulch includes an arpeggiator which plays continuously repeating arpeggios -- this is often a standard function on synthesizers. The arpeggiator cycles repeatedly through each note of a chord, possibly spanning multiple octaves. For example, given a chord C-E-G, the arpeggiator would play the related in a repeated sequence: C-E-G-E-C-E-G... Variations such as playing the notes in only an ascending or descending sequence are also possible. e.g. C-E-G-C-E-G-C... or G-E-C-G-E-C-G-...

attenuate / attenuation

The process of reducing (attenuating) the strength of a signal. In audio this usually means "turn the volume down" or "reduce the gain."

bandpass filter

A filter which passes only a certain band of frequencies. Frequencies outside the pass-band are rejected or attenuated.

bandwidth

The range of frequencies present in a signal, or the allowable range of frequencies supported by a transmission medium or filter.

See also broadband.

bar line

A vertical line on a music stave which delineates the end of one bar and the start of the next.

bar / bars

The space in between two bar lines. Indicates a full measure of time in the given time signature. For example, in 3/4 time, one full bar would consist of three quarter note (crotchet) beats of music.

See also bar line.

bit depth reduction / bit depth quantization

Bit depth reduction is a type of audio format conversion which reduces the amount of computer memory (disk space or RAM) consumed by an audio segment. A reduction of the the bit depth (see below) of digital audio is often performed when mastering audio for CD. Audio production is often performed using 24 or more bits per sample whereas CDs use 16 bit samples. Bit depth reduction may also be used arbitrarily as an audio effect (for example to simulate older sampling technologies which used 8 or 12 bits per sample) to create a kind of distortion. Quantization is one method of reducing bit depth where each sample is "snapped" to the nearest available value at the new bit depth -- this process may introduce audible distortion.

See also quantization.

bits / bit depth

Computers represent numbers using the binary counting system. Binary digits are called bits. A binary number with more bits (digits) can represent a larger range of numbers. For example, in the decimal system 2 digits support numbers from 0 to 99, while 4 digits support numbers from 0 to 9999. In binary, 2 digits support numbers from 0 to 11 (0 to 3 in decimal) while 8 bits support numbers from 0 to 11111111 (0 to 255 in decimal). Computers represent sounds using sequences of numbers (samples) -- assuming a fixed maximum loudness, the greater the number of bits used to represent each audio sample (called the bit depth), the greater the detail which can be represented. Thus, higher bit depths generally lead to clearer, less distorted sound, but also take up more computer memory.

broadband

An adjective used to describe a sound or signal with "wide bandwidth," which in audio terms equates to sounds with a range of different frequencies accross the whole audio spectrum, from high to low. Typical broadband sounds include human speech, white noise, or recordings of rock music with drums, bass and guitar. Conversely, narrowband sounds only contain frequencies in a narrow range, such as a single sine wave, birds chirping, or a low frequency rumble.

canon

A musical structure where the same melodic pattern is repeated by multiple voices, delayed by a fixed interval.

See also round.

carrier (signal)

A signal which gets modulated by a modulator. A carrier-modulator system is typically found in radio transmission, for example AM radio (where the amplitude of the carrier is modulated) or FM (where the frequency of the carrier is modulated). Both of these techniques have also been employed in audio synthesis.

See also modulation.

cents

A unit of measurement used to refer to the distance between two musical pitches. A cent is 1/100th of a semitone. Thus there are 1200 cents per octave. Relative pitch values can be positive or negative, for example middle C +50 cents is a quarter tone above middle C, middle C -100 cents is the B below middle C.

chord

Two or more musical pitches played simultaneously. Usually organized according to a system of harmony.

comb filter

A type of audio filter which passes or cuts narrow bands of frequencies spaced in a harmonic series -- i.e. the bands are centered on integer multiples of the lowest band. The filter is so named because its frequency response looks like like the equally spaced tines of a comb.

compression / dynamic range compression

The process of dynamic range compression reduces the dynamic range of a sound. This usually means that loud sounds are turned down so that they are not so loud. This is sometimes called "levelling" and can be used to make the loudness of a recording more consistent, which may make it easier to mix sounds together without unusually loud elements "poking out" inconsistently. The devices used to perform compression are called compressors. Compressors can produce a wide range of different sonic effects and are often used to apply distortion, rather than simply to change the dynamic range of a sound.

Further reading: http://en.wikipedia.org/wiki/Dynamic_range_compression

consonance

The musical concept of two or more notes played simultaneously sounding pleasing. For example, a fifth is said to be more consonant than a semitone. The opposite of consonance is dissonance.

Although there is some physical basis in consonance relating to beat-frequencies, there is no objective measure of what constitutes consonance and the term's exact definition has varied over the centuries. It is probably best to think of consonance and dissonance as the two extremes of a continuum.

See also dissonance.

cosine bell

A bell shaped curve described by the path of a single cycle of a cosine (or sine) waveform travelling from its minimum value up to its maximum value, and then back down to its minimum again.

CPU overload

When the computer's central processing unit (CPU) is unable to compute fast enough to achieve the desired result in the allocated time, the CPU is said to be overloaded. In audio processing applications this may mean you are trying to perform too much audio processing and the computer is not fast enough to keep up. The result is usually audio glitching and drop outs.

Note that despite the glitchy audio, the overload does not indicate a fault with the computer. It's not broken; it's just not coping.

cross-feedback

A feedback network which forms a cross or figure-eight loop (eg output from A feeds input of B, output of B feeds input of A. Stands in distinction to a simple feedback loop where the output of a process feeds back into its input.

See also feedback.

crossfader / crossfade

The process of crossfading involves mixing together two audio streams such that one fades out to silence as the other fades in. A crossfade refers to a single execution of the process of crossfading. A crossfader is a device (traditionally a mechanical slider on a DJ mixer) which allows you to perform the crossfading process by moving the slider from one end to the other to crossfade from one sound to another.

cycles per second

A unit of measure for frequency, usually used in audio systems to refer to the frequency of audio oscillators or low frequency oscillators. This unit is also known by the SI unit Hertz. It is used to refer to the number of times a vibrating system oscillates in a second.

See also Hertz.

Further reading: http://en.wikipedia.org/wiki/Cycle_per_second

dB (decibels)

A logarithmic unit of measure often used for representing measures of audio power. This is more meaningful than using linear voltage, as a logarithmic scale approximates more closely our perception of loudness. A change of approximately 6.02dB is more or less equivalent to a doubling in amplitude. However, it must be noted that there are many more subtleties to the way humans perceive loudness than can be captured by simply switching from a linear to logarithmic scale.

Decibels are relative units, thus always measured relative to a reference level. In a digital audio system such as AudioMulch the reference level is often dBFS (decibels relative to full scale -- the largest signal the audio hardware can reproduce). In this case 0dB corresponds to full scale, -6db is half of full scale and +6dB would be twice full scale.

Further reading: http://en.wikipedia.org/wiki/Decibel

delay / delay line

A delay line (sometimes simply delay) is a device which delays a signal (such as an audio signal) by a fixed or variable time. At least in principle, a delay line reproduces (outputs) its input acurately at some later time. Delay lines may have a fixed delay time, or the delay time may be varied (either by a user control, or using modulation such as a low frequency oscillator). A delay line may be tapped at multiple points to yield multiple outputs which are each a version of the input delayed by a different amount.

dissonance

The musical concept of two or more notes played simultaneously sounding displeasing. For example, a semitone is said to be more dissonant than a fifth. The opposite of dissonance is consonance.

See also consonance.

distortion

An effect where the quality of an audio signal is modified or degraded, it is commonly achieved by overdriving an electrical circuit beyond its normal range. There are as many types of distortion as there are circuits which produce it. Distortion always adds additional harmonics, which can make sounds more harsh and raspy. Some kinds of distortion, like tape saturation and tube amplifier distortion, are considered more "musical" than others, as they add subtle levels of additional harmonics, or interact in complex relationswith the amplitude of the source signal. More extreme distortion effects are commonly used with electric guitar.

doppler effect (doppler shift)

The doppler effect happens when the pitch of moving sounds is perceived as a shifting pitch that depends of the speed and direction of the sound source relative to the listener. This can be heard in everyday life as the sound of car and motorcycle engines appears to have a higher pitch as they approach and then shift lower as they pass and recede. This effect is due to the sound wavefronts compressing in front of the moving object, and expanding in its wake.

dynamic range

1. A measure of the variation in loudness within a signal or sound recording. For example, a piece of music which includes very quiet and very loud passages is said to have a wide dynamic range. The dynamic range of a sound can be reduced or limited using a compressor or limiter.

2. A measure of the variation in loudness that can by accommodated by a recording or transmission medium. Often dynamic range is limited by noise in the recording system -- sounds which are too quiet can not be captured because they are masked by noise.

dynamic range compressor

A dynamic range compressor (often just called a compressor, sometimes called a "leveller") acts to reduce the volume of loud signals. This creates a less dynamic sound that mixes more evenly with other sounds. Depending on the settings, a compressor can have other effects, such as making percussion sound more punchy.

envelope (loudness envelope)

The envelope of a sound refers to its overall loudness contour. Given an existing sound, we can impose a different envelope on it by changing its amplitude over time. Different sounds have different characteristic loudness envelopes. For example, a sound generated by striking a drum begins with an almost instantaneous rise to maximum loudness and then decays slowly. Other sounds may start quietly and slowly build to maximum amplitude, such as the sound of a passing car.

See also modulation.

envelope generator

A device which generates an envelope control signal.

See also ADSR envelope.

feedback

The term feedback refers to the (audible) results of a feedback loop. A feedback loop is a configuration of connections where the output of a process is fed back into it's input to be processed again (and again, and again, ad infinitum). This could be acoustical, for example where the output of a speaker is picked up by a microphone which is fed to a speaker. It could also be electronic or digital, for example, where the output of a digital delay is fed back to it's input to create repeating delays. Any process which has an input and an output may be subjected to feedback by connecting the output to the input.

The term feedback may also refer to the amount of gain or attenuation applied to the signal passing through the feedback loop -- applying more gain to a feedback loop typically results in greater audible feedback.

filter / filtering

An audio filter is a device which alters an audio signal, usually by boosting or cutting certain frequencies. Filters which make fine adjustments are often called equalization filters (abbreviated EQ). Others have more extreme effects such as removing whole areas of the frequency spectrum (lowpass, highpass, bandpass). Some filters only alter the phase of the signal (see all-pass filter).

flanger

A device which imposes a type of audio effect called flanging on an audio signal. Flanging was originally achieved by playing the same sound on multiple tape recorders and varying their relative speed by leaning on the tape reel flange. More commonly, a flanger is implemented using an electronic or digital circuit which approximates the sound of the tape process.

frequency

1. A quantity describing the rate at which a periodic cycle occurs, such as the rate of vibration of a membrane or string, or other sound producing mechanism. Frequency is measured using units of Hertz, the number of cycles per second, abbreviated Hz. E.g. "The oscillator generates a sine wave with a frequency of 200Hz." Sometimes kHz (1000's of cycles per second) is used. The human auditory system is generally said to be able to detect frequencies from 20Hz to 20Khz, although adults usually can't hear much above 16kHz.

2. The terms frequency and frequencies are sometimes used to refer to specific sounds, or components of a sound. For example, an individual sine tone may be referred to as a 'frequency'.

3. Similarly to (2) frequency and frequencies may be used to refer to a specific region of the sound spectrum. For example, "we boosted all frequencies below 50Hz to achieve the brown note." Less specifically, one might refer to "high frequencies," "low frequencies," "mid-range" frequencies.

fundamental (frequency)

The lowest frequency in a harmonic tone. If all harmonics are present in a tone then this is also the spacing between adjacent harmonics. This is also the frequency which determines the musical pitch of a tone. Humans are able to perceive the fundamental frequency (or pitch) of a sound even the actual frequency is not physically present. For example, a small speaker cannot physically reproduce the fundamental frequency of a low bass instrument, but we may still perceive it to be present from the higher harmonics.

gain

In simple terms, applying gain turns the volume up or down. Positive gain increases the amplitude of an audio signal. A negative amount of gain decreases the amplitude. "Unity gain" means not changing the amplitude of the signal.

gate / gated / gating

The process of repeatedly switching an audio signal on and off, possibly in a rhythmic pattern. Gating can be thought of as applying an amplitude envelope with a fast attack and release time. Gating may be applied manually using a switch, by programming a rhythmic pattern or it may be applied automatically, based on the level of a sound (as in a noise gate).

See also noise gate.

glissando

A smooth slide from one pitch to another.

granular synthesis / granulation

A sound synthesis technique where an aggregate sound mass is created by mixing together many small grains of sound. When the grains of sound are created by cutting and splicing together small fragments of a pre-existing sound source, the process is referred to as granulation.

harmonic

A single frequency component in a harmonic series. Harmonics are, by definition, sine waves. Sometimes called a partial.

See also harmonic series, harmonics.

harmonics

A group or subset of individual harmonic components in a harmonic series. This is often qualified, as in odd-harmonics, even-harmonics, upper harmonics. Sometimes used to draw a distinction with the fundamental frequency of a harmonic series.

See also harmonic series, harmonic.

harmonic series

A group of sine wave components whose frequencies are related by being integer multiples of a fundamental frequency. Vibrating systems such as strings and tubes tend to produce tones comprising frequencies which are harmonically related due to the physical constraints of the system -- a string only supports vibrations which fit exactly along the length of the string (one times its length, half its length, one third its length and so on).

Further reading: http://en.wikipedia.org/wiki/Harmonic_series_(music)

high-hat

A mechanism of two cymbals mounted on a single pole often found as part of a drum kit. The cymbals can sit apart ("open") or can be temporarily clamped together ("closed") by the player using a foot pedal. When struck, the high-hat makes different sounds when open or closed. It is possible to strike it while open and to then close it to gate the sound while it is ringing out. Playing a high-hat often involves intricate patterns of opening and closing the cymbals while striking them with drum sticks.

high shelf filter

A filter that allows adjustment of the gain of frequencies above a specific shelf freqency. Shelving filters can boost or cut frequencies.

See also low shelf filter.

Further reading: http://en.wikipedia.org/wiki/Filter_design

Hz (Hertz)

The SI unit for frequency. Frequency, measured in Hertz, is the number of times a periodic event such as an audio vibration occurs in one second.

See also frequency.

Further reading: http://en.wikipedia.org/wiki/Hertz

inharmonic spectrum

A sound spectrum which includes partials which are not harmonically related. For example, most bells and gongs produce sounds with resonances which are not harmonically related.

See also harmonic.

Further reading: http://en.wikipedia.org/wiki/Inharmonicity

inskip

Offset from the start of a sound file or sound sample. The distance from the start of a sound file to a specific location within the file.

LFO (low frequency oscillator)

An oscillator which generates a slowly varying waveform which is often used to modulate the amplitude, pitch or filter settings in a synthesizer. Here low frequency refers to frequencies below the audio rate (less than approximately 20Hz). Instead of hearing the oscillator as a pitch (as in an audio rate oscillator), a low frequency oscillator is usually heard as a repeated rhythmic modulation of synthesis parameters.

low bit rate

An attribute of a digital audio stream (or other digital stream) which uses a small number of bits per second to represent a signal. Usually this corresponds to low audio fidelity. Most often this is used when referring to compressed audio (such as mp3 audio) but it also applies to linear audio of either a low sample rate, low bit depth, or both.

See also bit depth, sample rate.

low shelf filter

A filter that allows adjustment of the gain of frequencies below a specific shelf freqency. Shelving filters can boost or cut frequencies.

See also high shelf filter.

Further reading: http://en.wikipedia.org/wiki/Filter_design

low pass filter

A low pass filter is a filter which passes low frequencies, and consequently elminates high frequencies. The cutoff frequency is the point where the filter transitions from pass band to stop band. Usually practical filters don't exhibit a perfect pass/stop frequency response, there is a gradual transition from passband to stop band, and attenuation in the stop band is not complete but rather significantly stronger than in the pass band. For example, a common characteristic of musical lowpass filters is that they reduce gain by 24dB for each octave above the cutoff frequency.

low pass resonant filter

A type of low pass filter (see above) which exhibits a resonant peak around the cutoff frequency. That is, it boosts frequencies around the cutoff frequency. This can create a nasal or formant-like quality to filtered sounds, and is a key characteristic of the classic "analog filter sweep" sound. Lowpass filters with resonance typically allow you to adjust the amount of resonance, which is sometimes called "Q" -- the engineering term for resonance.

melody

A sequence of notes, heard one after the other, consisting of a variety of pitches. Rhythm also plays a part in defining the character of a melody.

See also rhythm.

midrange filter

A filter which operates on frequencies in the middle of the audible frequency spectrum. Approximately from 100Hz to 2000Hz.

See also filter.

mixer

A device which combines (mixes) multiple audio streams into a single stream. Usually a mixer allows the relative amplitude (loudness) of each stream to be adjusted separately and may allow other changes to each stream. For example, a stereo mixer might allow each stream to be panned separately.

modulation / modulator

To "modulate" means to change or vary. Hence, modulation is the process of changing something and a "modulator" is that which effects the change. In sound synthesis, modulation usually means to apply an ongoing time-varying change to some sound parameter such as frequency, amplitude, etc. Most often the modulation is performed by an ongoing or automatic process, for example, an LFO (low frequency oscillator) or envelope follower -- such modulation sources are referred to as "modulators". ADSR envelopes are another common form of modulation sources.

mono / monophonic / monaural

Mono is an abbreviation of monophonic or monaural. The term has two uses, one in conjunction with spatial (stereo or multichannel surround audio) and another in relation to instruments capable of only producing one tone at a time (the human voice for example, as opposed to a polyphonic instrument such as a pipe organ).

A single audio source. For example, audio emanating from a single loudspeaker. Or audio passing through a single wire or connection. Usually used in contrast to stereo or multichannel.

Further reading: http://en.wikipedia.org/wiki/Monaural

multichannel

An audio system which maintains audio in multiple discrete channels. This is usually used to mean more than one (mono) or two (stereo) channels. For example, a multichannel speaker system, often used in surround sound reproduction, employs a number of speakers, each fed with a different audio signal. A multi-channel audio interface supports output of more than two audio signals. A multi-channel sound file contains more than two audio channels.

multitrack

A recording or playback system where multiple tracks or channels of independent audio are recorded and played back. Often multitrack systems are used in recording studios, where each instrument in a performance is recorded on a separate track -- sometimes multiple tracks are used to capture different microphones capturing a single instrument. Multitrack is different from multichannel in that a multitrack recording may be (and often is) mixed down to mono or stereo -- in this sense it is a recording technology, whereas multichannel is usually used to describe a delivery system or a capability to deal with many channels of independent audio.

noise gate

A noise gate (sometimes just "gate") is an audio processing device originally intended to suppress background noise. It opens and closes, allowing louder sounds through while muting or reducing the gain of quieter sounds. Traditionally, it has been used to mute out background noise when foreground sound is not present. Some noise gates can be configured as "duckers" which mute loud sounds. In a side chain configuration a ducker can be made to turn down one sound (such as background music) while another sound is present (such as a voice over).

See also side chain.

Further reading: http://en.wikipedia.org/wiki/Noise_gate

normalize

The process of adjusting the volume of an audio signal or waveform to a specified level. Usually this would mean adjusting the volume of a waveform so its maximum level is the maximum level representable in the system, perhaps with a small safety margin, for example normalizing the peak level to 3db below maximum.

octave

A musical term for a distance between pitches which corresponds to a doubling (or halving) in frequency. Musical notes an octave apart belong to the same pitch class (for example, the C above middle C is an octave above middle-C). There are 7 notes in a diatonic scale, the eighth being the note an octave above the first -- this is the origin of the "oct" in the word "octave", as the Latin name for the number 8 is "octo".

oscillator

A device which repeatedly performs the same action is said to oscillate and is called an oscillator. An audio oscillator produces a repeated audio waveform, which results in a tone of fixed pitch and consistent timbre. Audio oscillators may vary in frequency, which alters their pitch, and may produce a variety of waveforms, which determines their timbre.

pan / panning

Derived from the word "panorama," panning refers to the process of placing a sound at a specific location in a spatial mix. In a stereo environment, this means placing the sound somewhere between the left and right loud speakers. Panning is usually performed with a pan control, such as a knob, which allows you to dial in the location.

parametric equalizer

A type of filter used to adjust an audio signal. An equalizer allows the adjustment of certain frequency bands within an audio signal -- for example a simple tone control found on a hi fi system which allows boosting or reducing high and low frequencies is a type of equalizer. A parametric equalizer is a type of equalizer allowing more precise control over the frequencies which are adjusted by the equalizer. A typical parametric equalizer allows tuning in to specific frequency bands by adjusting cutoff frequencies and bandwidths -- the ability to tune in to these frequencies (the parameters) is what makes a parametric equalizer parametric.

phase cancellation

When two audio signals are mixed, some frequencies may coincide and reinforce or cancel out each other. Phase cancellation refers to cancellation effects due to two frequencies being the same but out-of-phase such that one partially or totally cancels the other. The phase difference may be accidental, due to the time delay effects of capturing the same source using multiple microphones. It can also be deliberate, achieved by splitting the signal into two paths, and processing each path differently before recombining both signals. Phase cancellation is most often an undesirable effect which reduces the strength of an audio signal and weakens some harmonics making the sound sound "thin".

See also phasing.

phase inversion

The process of inverting the polarity of an audio signal. Such a signal is said to be "out of phase" with the non-inverted original. In this sense, phase inversion is only relevant when there is some "in phase" signal to compare with. Mixing an audio signal with a phase-inverted copy results in cancellation and therefore silence. The effect of phase inversion is perhaps most commonly encountered when one miswires one channel of a stereo speaker system so that the wires are swapped on one speaker. This creates a kind of stereo-wide sensation as sounds which should be heard as common between the two speakers are completely out of phase and are heard as though they were emanating "beyond the speakers".

phaser

An audio effect which filters and recombines multiple streams of its input in such a way as to produce sweeping resonant peaks or swooshing sounds. Often the rate and depth of the effect can be controlled.

phasing

1. An audible effect heard as a high-frequency swishing sound which results from phase cancellation. The effect may arise from mixing back two very similar audio sources. It can result from mixing audio recorded from multiple microphones at different locations, from mixing sounds processed by different effects, or by special audio processing effects such as phasers and flangers which mix together sounds filtered in different ways to create sweeping effects.

2. The outcome of playing back multiple streams of repeated or looped rhythmic or time structured audio (such as speech) of different lengths such that at each length the streams are heard against each other at different phases. This kind of phasing can produce slowly changing patterns (if loop lengths are very close) or more chaotic variations (if the phases at each repeat vary dramatically). Some music, such as that by Minimalist composer Steve Reich, makes heavy use of phasing as a compositional technique.

See also phase cancellation, phaser, flanger.

pitch

1) A musical term used to describe the fundamental frequency of a musical tone within a particular tuning system. Usually the 12-tone tuning system is used where each octave is broken up in to 12 equally spaced intervals and these are referred to by the pitch classes A, A#/Bb, B, C, C#/Db, D, D#/Eb, E, F, F#/Gb, G, G#/Ab.

2) A descriptive term used to refer to the relative frequency of an audio signal (e.g. "a high pitched sound").

pulsar synthesis

A form of sound synthesis which generates a train of pulses with regular period. Depending on the rate of pulse generation, the pulse train may be perceived as a regular rhythm, or at faster rates, as a pitched tone. The ability of pulsar synthesis to span these two perceptual domains is one of its key features. At is simplest, the pulses may be impulses (clicks) repeated at a certain rate, as in a classical analog pulse generator used in early electronic music. More commonly, the pulses are small grains of sampled sound, making pulsar synthesis a variation on granular synthesis.

Further reading: http://en.wikipedia.org/wiki/Pulse_generator

pulse train

A repeated sequence of pulses. In this context, a pulse is a short click-like sound, usually approximating a dirac-pulse, which is a click with energy at all frequencies. Unlike most audio waveforms, whose higher harmonics are weaker (ie the sound has more low frequency energy than high frequency) a pulse train is a periodic (pitched) sound all of whose harmonics have equal amplitude.

quantization

The process of adjusting something (such as the time of events) such that it snaps to a regular grid. For example, it is possible to quantize pitches to conform to a musical scale, or perhaps more commonly, to quantize rhythmic events such that the conform to a regular pulse. Quantization need not be all-or-nothing, it's possible to have partial quantization where events are moved some way towards a regular pulse.

The term quantization can also refer to a process of bit depth reduction.

See also bit depth reduction.

resonance

A region in the frequency response of a system where frequencies are amplified or reinforced. This may be heard as a "ringing" sound, where the system (an audio filter, and effect, a bell, a room) continues to sound even when no energy is fed into the system. Acoustic systems such as bells resonate at discernible pitches; others such as the bodies of guitars emphasize a broad range of frequencies.

See also resonant filter.

resonant filter

A filter which exhibits resonance at a certain frequency or frequencies. Often this refers to a low pass resonant filter, although the term covers any filter which rings or resonates. Other examples combine filters with feedback (as found in AudioMulch's 5Combs contraption) or more complex filters which simulate resonant physical structures such as tubes or plates.

See also low pass resonant filter.

reverberation

The resonant characteristic of rooms and other large architectural spaces. Reverberation is the result of sound reflecting off the many surfaces of the room and being refracted and delayed as it travels through the air. The sound reaches the listener through many of these different paths as it bounces around the room; the result is a delayed and smeared effect, which is perceived as a sense of the size and character of a space.

The sound of reverberation is heard when clapping in a particular space, perhaps most easily in a big church or cathedral. Rooms designed for musical performance are often acoustically treated and proportioned to create reverberation which enhances the musical experience. Reverberation may be simulated by artificial means using a reverb processor. This effect is often used in order to create a sense of depth in electronic and recorded music without necessarily seeking to simulate a real space.

rhythm / rhythmic

Rhythm in its most basic sense means the duration of a note. Rhythm is part of a complex system of musical time which, in Western culture, also involves time signatures, beats, stresses, and tempo/speed. Rhythm is typically used to create beats and patterns, and the rhythm in a piece of music is often made up of a range of durations. In Western music many rhythmic values are based around subdivisions of time, and tend to be related to one another by simple mathematical equations such as division or multiplication by two (in duple time signatures such as 2/4 and 4/4) or three (in compound time signatures such as 3/8 and 6/8). Rhythms and durations are subject to a number of rules in traditional Western music systems, including how particular rhythmic values should be grouped, and which beats or notes should be accented in a particular time signature.

See also time signature.

ring modulation

An electronic effect which involves modulating one signal with another. It is equivalent to multiplying the instantaneous amplitude of one signal by another, which produces a resulting sound with harmonics at the sums and differences between all frequencies in the two input sounds. Often ring modulation is performed with one signal being a sine wave. This results in two shifted copies of all of the frequencies in the non-sine wave signal: one set (the sums) shifted upwards by the frequency of the sine wave, the other set (the difference) being shifted relative to the difference between the source and the sine wave.

round

A simple form of musical canon, usually sung. It is often for three or more voices, each singing the same melody in unison or an octave apart. Rhythmically, the entry point of each singer is offset, usually by a regular interval of time, so that each new entry of the melody overlaps with the others to form harmonies. A well known round is "Frere Jacques".

sample

1. A captured segment of digital audio. Usually refers to a recording of an individual sound (such as a drum hit) or a portion of a prexisting piece of recorded music.

2. An individual number, which forms part of a sequence of numbers which together represent the changing values of a digital audio signal over time.

See sample rate.

sampling

The process of creating recorded digital audio segments from pre-existing acoustic or recorded music sources. This process is sometimes part of a cultural practice where recognisable or obscure segments of other people's music is sampled and used to make new music through a process of bricolage.

Further reading: http://en.wikipedia.org/wiki/Sampling_(music)

sample inskip

See inskip.

sample rate

Computers represent digital audio as a sequence of numbers called samples. Each sample represents a voltage level in an electronic audio system, usually measured from a microphone, or used to drive a loud speaker. These voltages correlate to changing air pressure levels as sound travels through the air. The sample rate or sampling rate then, is the rate at which voltages are measured to create the sequence of sample values. CD audio for example, is sampled at a rate of 44100 samples per second.

See also sample.

sample rate decimation

The process of reducing the sample rate by discarding samples. For example, the sample rate can be (naively) reduced by a factor of two by dropping every second sample. Note that decimation performed in this manner introduces aliasing distortion (see below), more often careful filtering operations are used to avoid negative effects of sample rate conversion.

sample rate decimation aliasing

The distortion which results from performing sample rate conversion using the simple process of discarding samples to reduce the sample rate. When aliasing occurs, audible frequencies present in the original signal which were above half the new sample rate are reflected across this half-sample-rate boundary, and will be heard as (usually undesirable) distortion frequencies. This phenomenon is the digital audio correlate of wagon wheels being seen to spin backwards in movies because the film frame rate is not fast enough to capture their forward motion.

sample rate reduction

The process of reducing the sample rate of a digital audio signal.

See also sample rate, sample rate decimation.

sawtooth wave

A waveform which looks like a saw blade: one vertical edge and another ramped edge. Such waveforms are commonly used in analog synthesizer oscillators. Sawtooth waves have both odd and even harmonics which decay slowly, producing a distinctive bright sound which is suitable for filtering with a lowpass filter.

semiquaver

A note that is one sixteenth of a whole note in duration. In a bar of 4/4 time, there are sixteen sixteenth notes (semiquavers) in a bar. In traditional Western musical notation, the note head is filled in black and has two tails (flags) attached to its stem. It is half the duration of an eighth note (quaver).

See also time signature, rhythm.

semitone

The difference between two adjacent musical pitches in the western diatonic system. This is the distance between two adjacent notes on a piano keyboard (when using both white and black keys) or two adjacent frets on a guitar.

sequencing

The process of composing or specifying a musical sequence comprising rhythm and pitch, usually of musical notes and chords, although it may also include specification of dynamics, timbral modulations and so on. Sequencing is often accomplished by specialized software called a sequencer, or it may be accomplished using sequencer-like functionality embedded in other software as is the case in AudioMulch.

sidebands

The resultant harmonic components generated by some modulation synthesis techniques such as amplitude modulation.

side chain

A separate chain of audio processing or filtering which sits to the side of a main processing chain. Commonly this is used in audio processes where the processing is controlled by the audio input. Rather than use the same audio as the source to be processed, and the source to control the process, a separate side chain may be used to provide the control signal. For example, a kick drum may be used to control a nose gate which processes a bass guitar. In some cases, the side chain may be a separately filtered version of the source signal -- for example, applying separate filtering to the control signal used for a compressor to make the compressor more or less sensitive to transients.

(audio) signal generator

The electronic equivalent of a sound producing object. Signal is a general term for a fluctuating electrical voltage and current. In audio systems, signals often represent acoustical fluctuations captured by microphones and/or reproduced by speakers. A signal generator then, is an electronic (or digital) circuit or system which produces fluctuating voltages.

sign bit

Digital systems represent numbers (and hence audio signals) using the binary counting system, where digits can be either 0 or 1. Such binary digits are called "bits". One way of representing negative numbers in a binary counting system is to use a bit to indicate whether the number is positive or negative - such a bit is called the "sign bit."

sine wave

A waveform distinguished by its lack of harmonics. It contains only a single frequency component. A sine wave is curved, and its shape is related to the motion on a single axis of a point travelling around the perimeter of a circle.

sinusoidal

In the shape of a sine wave.

See also sine wave.

sixteenth note (semiquaver)

See semiquaver.

spatializer

A device which "spatializes" sound. The process of spatialzing involves imbuing a sound with spatial properties, which often but not always includes motion in space. A spatializer may create different illusions: that a sound source is placed in space between two or more speakers, that it is moving between the speakers or that they are placed at different depths, seeming to be closer or further away. There are many different techniques for creating a sense of sound in space; some of them employ physical properties of sound (e.g. distant sounds are quieter, doppler shift of moving sounds) while others address specifics of the human auditory system (e.g. stereo panning between two speakers, arrival time differences between the two ears).

square wave

A waveform with two constant level states of equal length, each of equal level but opposite polarity. Such waveforms have only odd harmonics and as a result have a "hollow" sound like a clarinet or other woodwind instrument.

stereo

A method of audio recording and reproduction where two distinct audio channels (left and right) are used to reproduce a lateral sound field using either two loud speakers or headphones. The term stereo may be used to denote a device using 2 audio channels as opposed to 1 (mono), or more than 2 (quad, surround, octaphonic etc).

summed / summing

The process of adding two or more things together. In audio this refers to mixing two or more audio signals together at equal strength.

(audio) synthesis

The process of creating audio using one or more mechanisms (electronic or software based) in combination. For example, additive synthesis mixes sinewaves, subtractive synthesis combines oscillators and filters.

synthesize(r)

A device which performs audio synthesis.

tempo

The speed of the beat in rhythmic music. The plural of tempo is 'tempi'. In music software, tempi are usually specified as a rate of "beats per minute".

timbre

A term used to refer to the quality of a sound not covered by notions of loudness or pitch. The term has a number of meanings but a common ones are tone-color. It may be used to discuss the tone colour of a specific instrument (e.g. "that violin has a different timbre to the other one") or to distinguish between different sound categories (e.g. "a church bell has a completely different timbre to the sound of waves crashing at the seaside").

Further reading: http://en.wikipedia.org/wiki/Timbre

time signature

A numerical sign, usually placed near the start of a piece of music - after the clef and key signature, if any - that represents the timing or metre of the music. It can also be placed at any other point in a piece of music. A time signature usually consists of two numbers, one one top of the other. The upper number tells you how many beats there are in each bar, while the lower number indicates the type of beat. The lower number represents the duration of each beat as a fraction of a whole note, or semibreve. Therefore, in the time signature 3/8, the bottom number indicates that the beat is 1/8 the length of a whole note, which is an eighth note (or quaver). The upper number, 3, indicates that there are three of these eighth note (quaver) beats per bar.

See also bar, rhythm.

transpose / transposition

The musical process of transposition, transposing (verb) a melody or sound means changing all of its pitches from one pitch level to another. A transposition (noun) refers to an instance of a melody at a particular pitch level. Chromatic transposition is performed by shifting all pitches up or down by the same interval. The interval may be expressed in semitones or (for microtonal transpositions) in cents. Diatonic transposition moves pitches by scale degree rather than chromatic interval -- resulting in a melody which is in the same key as the untransposed original.

Further reading: http://en.wikipedia.org/wiki/Transposition_(music)

trigger

A instantaneous signal to initiate an action. In sound synthesis this could mean to cause an envelope to begin, or to cause a sound to be played. Unlike a musical note, which has a start, a duration and an end, a trigger is an instantaneous event.

waveshaping

A sound synthesis technique where an input waveform is warped or bent by subjecting it to a waveshaping function which specifies for each input level a corresponding output level. Waveshaping is a form of distortion which always adds additional harmonics to a sound. Methods for computing wave shaping functions exist which allow exact specification of the added harmonics given a certain input (for example the Chebyshev waveshaping technique used in AudioMulch's Shaper supports specifying output harmonics for a sine wave input). Like distortion, waveshaping is a nonlinear process which means the sonic results can be highly dependent on the nature and loudness of the input sound.

white noise

A type of noise which on average contains equal energy at all frequencies.

Further reading: http://en.wikipedia.org/wiki/White_noise