Glossary of Electronic Music Terms
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z 


A ADC: See analog-to-digital converter.
A/D converter: See analog-to-digital converter.
ADPCM: Adaptive delta pulse code modulation. An audio compression algorithm for digital audio based on describing level differences between adjacent samples. 
ADSR: Attack/decay/sustain/release, the four segments of a common type of synthesizer envelope. The controls for these four parameters determine the duration (or in the case of sustain, the height) of the segments of the envelope. See envelope.
aftertouch: A type of control data generated by pressing down on one or more keys on a synthesizer keyboard after they have reached and are resting on the keybed. See channel pressure, poly pressure.
AIFF: Audio interchange file format. A common Macintosh audio file format. It can be mono or stereo, at sampling rates up to 48kHz. AIFF files are QuickTime compatible.
algorithm: A set of procedures designed to accomplish something. In the case of computer software, the procedures may appear to the user as a configuration of software components -- for example, an arrangement of operators in a Yamaha DX-series synthesizer -- or as an element (such as a reverb algorithm) that performs specific operations on the signal.
algorithmic composition: A type of composition in which the large outlines of the piece, or the procedures to be used in generating it, are determined by the human composer while some of the details, such as notes or rhythms, are created by a computer program using algorithmic processes.
aliasing: Undesired frequencies that are produced when harmonic components within the audio signal being sampled by a digital recording device or generated within a digital sound source lie above the Nyquist frequency. Aliasing differs from some other types of noise in that its pitch changes radically when the pitch of the intended sound changes. See Nyquist frequency.
all-notes-off: A MIDI command, recognized by some but not all synthesizers and sound modules, that causes any notes that are currently sounding to be shut off. The panic button on a synth or sequencer usually transmits all-notes-off messages on all 16 MIDI channels.
amplitude: The amount of a signal. Amplitude is measured by determining the amount of fluctuation in air pressure (of a sound), voltage (of an electrical signal), or numerical data (in a digital application). When the signal is in the audio range, amplitude is perceived as loudness.
analog: Capable of exhibiting continuous fluctuations. In an analog audio system, fluctuations in voltage correspond in a one-to-one fashion with (that is, are analogous to) the fluctuations in air pressure at the audio input or output. In an analog synthesizer, such parameters as oscillator pitch and LFO speed are typically controlled by analog control voltages rather than by digital data, and the audio signal is also an analog voltage. Compare with digital.
analog-to-digital (A/D) converter (ADC): A device that changes the continuous fluctuations in voltage from an analog device (such as a microphone) into digital information that can be stored or processed in a sampler, digital signal processor, or digital recording device.
attack: The first part of the sound of a note. In a synthesizer envelope, the attack segment is the segment during which the envelope rises from its initial value (usually zero) to the attack level (often the maximum level for the envelope) at a rate determined by the attack time parameter.
attenuator: A potentiometer (pot) that is used to lower the amplitude of the signal passing through it. The amplitude can usually be set to any value between full (no attenuation) and zero (infinite attenuation). Pots can be either rotary or linear (sliders), and can be either hardware or "virtual sliders" on a computer screen.
auto-correct: See quantization.
B
bandwidth: The available "opening" through which information can pass. In audio, the bandwidth of a device is the portion of the frequency spectrum that it can handle without significant degradation. In digital communications, the bandwidth is the amount of data that can be transmitted in a given period of time.
bank: (1) A set of patches. (2) Any related set of items, e.g., a filter bank (a set of filters that work together to process a single signal).
baud rate: Informally, the number of bits of computer information transmitted per second. MIDI transmissions have a baud rate of 31,250 (31.25 kilobaud), while modems typically have a much lower rate of 2,400, 9,600, or 14,400 baud.
bend: To change pitch in a continuous sliding manner, usually using a pitch-bend wheel or lever. See pitch-bend.
bit: The smallest possible unit of digital information, numerically either a 1 or a 0. Digital audio is encoded in words that are usually eight, 12, or 16 bits long (the bit resolution). Each added bit represents a theoretical improvement of about 6dB in the signal-to-noise ratio. 
bpm: Beats per minute. The usual measurement of tempo.
brick-wall filter: A lowpass filter at the input of an analog-to-digital converter, used to prevent frequencies above the Nyquist limit from being encoded by the converter. See Nyquist frequency, aliasing.
buffer: An area of memory, used for recording or editing data before it is stored in a more permanent form.
bulk dump: See data dump.
byte: A group of eight bits. (MIDI bytes consist of ten bits because each byte includes a start bit and a stop bit, with eight bits in the middle to convey information.)
C
card: (1) A plug-in memory device. RAM cards, which require an internal battery, can be used for storing user data, while ROM cards, which have no battery, can only be used for reading the data recorded on them by the manufacturer. (2) A circuit board that plugs into a slot in a computer.
carrier: A signal that is being modulated by some other signal, as in FM synthesis.
CD-ROM: Compact disc read-only memory. A compact disc format that can store data other than just standard CD audio. Many programs, sound sample libraries, and graphics are distributed on CD-ROM because each CD can store hundreds of megabytes of information, yet costs about the same to manufacture as a floppy disk, which only stores about 1 megabyte. See ROM.
cent: The smallest conventional unit of pitch deviation. One hundred cents equal one half-step.
channel: An electrical signal path. In analog audio (such as a mixer), each channel consists of separate wired components. In the digital domain, channels may share wiring, and are kept separate through logical operations. MIDI provides definitions for 16 channels, which transmit not audio signals but digital control signals for triggering synthesizers and other devices.
channel pressure: A type of MIDI control message that is applied equally to all of the notes on a given channel; the opposite of poly pressure, in which each MIDI note has its own pressure value. Also called aftertouch, channel pressure is generated on keyboard instruments by pressing down on a key or keys while holding them down. See aftertouch, poly pressure.
chorusing: A type of signal processing. In chorusing, a time-delayed or detuned copy of a signal is mixed with the original signal. The mixing process changes the relative strengths and phase relationships of the overtones to create a fatter, more animated sound. The simplest way to achieve chorusing is to detune one synthesizer oscillator from another to produce a slow beating between them.
clangorous: Containing partials that are not part of the natural harmonic series. Clangorous tones often sound bell-like.
clock: Any of several types of timing control devices, or the periodic signals that they generate. A sequencer's internal clock is always set to some number of pulses per quarter-note (ppq), and this setting is one of the main factors that determine how precisely the sequencer can record time-dependent information. The actual clock speed is usually determined by the beats-per-minute setting. See ppq, bpm, MIDI clock.
clock resolution The precision (measured in ppq) with which a sequencer can encode time-based information.
companding: A type of signal processing in which the signal is compressed on input and expanded back to its original form on output. Digital companding allows a device to achieve a greater apparent dynamic range with fewer bits per sample word.
compression: (1) The process of reducing the amplitude range of an audio signal by reducing the peaks and bringing up the low levels. (2) The process of reducing a data file in size, often by noting patterns in the data and summarizing them. Some types of audio data compression are "lossy," meaning the quality of the audio is reduced.
continuous controller: A type of MIDI channel message that allows control changes to be made in notes that are currently sounding. See controller.
controller: (1) Any device -- for example, a keyboard, wind synth controller, or pitch-bend lever -- capable of producing a change in some aspect of a sound by altering the action of some other device. (2) Any of the defined MIDI data types used for controlling the ongoing quality of a sustaining tone. Strictly speaking, MIDI continuous controllers are numbered from 0 to 122; in many synthesizers, the controller data category is more loosely defined to include pitch-bend and aftertouch data.
crossfade looping: A sample-editing feature found in many samplers and most sample-editing software, in which some portion of the data at the beginning of a loop is mixed with some portion of the data at the end of the same loop, so as to produce a smoother transition between the end and the beginning when the loop plays. 
cross-switching: A velocity threshold effect in a synthesizer in which one sound is triggered at low velocities and another at high velocities, with an abrupt transition between the two. If the transition is smooth rather than abrupt, the effect is called crossfading rather than cross-switching. Cross-switching can also be initiated from a footswitch, LFO, or some other controller. Also called velocity switching.
cutoff frequency: The point in the frequency spectrum beyond which a synthesizer's filter attenuates the audio signal being sent through it.
D
DAC: See digital-to-analog converter.
data dump: A packet of memory contents being transmitted from place to place (usually in the form of MIDI system-exclusive data) or stored to a RAM card.
daughterboard: A small circuit board that can be attached to a larger one (the motherboard), giving it new capabilities. For example, some companies manufacture daughterboards that add sampled sounds to soundcards that previously could only synthesize sounds via FM.
dB: See decibel.
decay: The second of the four segments of a typical ADSR envelope. The decay control determines the amount of time it takes for the envelope to fall from the peak reached at the end of the attack segment to the sustain level. See ADSR.
decibel: A unit of measurement used to indicate audio power level. Technically, a decibel is a logarithmic ratio of two numbers, which means that there is no such thing as a dB measurement of a single signal. In order to measure a signal in dB, you need to know what level it is referenced to. Commonly used reference levels are indicated by such symbols as dBm, dBV, and dBu.
delay: (1) The first stage of a five-stage DADSR envelope, which delays the beginning of the envelope's attack segment. (2) A control function that allows one of the elements in a layered sound to start later than another element. (3) A signal processor, used for flanging, doubling, and echo, that holds its input for some period of time before passing it to the output, or the algorithm within a signal processor that creates delay.
detune: Noun: A control that allows one oscillator to sound a slightly different pitch than another. Verb: To change the pitch of one oscillator relative to another, producing a fuller sound.
digital: Using computer-type binary arithmetic operations. Digital music equipment uses microprocessors to store, retrieve, and manipulate information about sound in the form of numbers, and typically divides potentially continuous fluctuations in value (such as amplitude or pitch) into discrete quantized steps. Compare with analog.
digital-to-analog converter (DAC): A device that changes the sample words put out by a digital audio device into analog fluctuations in voltage that can be sent to a mixer or amplifier. All digital synthesizers, samplers, and effects devices have DACs (rhymes with fax) at their outputs to create audio signals.
DirectX: This Microsoft Windows API was designed to provide software developers with direct access to low-level functions on PC peripherals. Before DirectX, programmers usually opted for the DOS environment, which was free of the limited multimedia feature set that characterized Windows for many years.
download: To transfer a file from another computer into your own. Often done by modem. See modem.
dry: Consisting entirely of the original, unprocessed sound. The output of an effects device is 100% dry when only the input signal is being heard, with none of the effects created by the processor itself. Compare with wet.
DSP: Digital signal processing. Broadly speaking, all changes in sound that are produced within a digital audio device, other than changes caused by simple cutting and pasting of sections of a waveform, are created through DSP. A digital reverb is a typical DSP device.
dump: see data dump.
dynamic voice allocation: A system found on many multitimbral synthesizers and samplers that allows voice channels to be reassigned automatically to play different notes (often with different sounds) whenever required by the musical input from the keyboard or MIDI.
E
early reflections: A reverb algorithm whose output consists of a number of closely spaced discrete echoes, designed to mimic the bouncing of sound off of nearby walls in an acoustic space.
echo: A discrete repetition of a sound, as opposed to reverberation, which is a continuous wash of closely spaced, non-discrete echoing sound. See delay (3), reverb.
edit buffer: An area of memory used for making changes in the current patch. Usually the contents of the edit buffer will be lost when the instrument is switched off; a write operation is required to move the data to a more permanent area of memory for long-term storage.
editor/librarian: A piece of computer software that allows the user to load and store patches and banks of patches (the librarian) and edit parameters (the editor).
effects: Any form of audio signal processing -- reverb, delay, chorusing, etc.
envelope: A shape that changes as a function of time. The shape of a synthesizer's envelope is controlled by a set of rate (or time) and level parameters. The envelope is a control signal that can be applied to various aspects of a synth sound, such as pitch, filter cutoff frequency, and overall amplitude. Usually, each note has its own envelope(s).
envelope generator: A device that generates an envelope. Also known as a contour generator or transient generator, because the envelope is a contour (shape) that is used to create some of the transient (changing) characteristics of the sound. See ADSR, envelope.
envelope tracking: A function (also called keyboard tracking, key follow, and keyboard rate scaling) that changes the length of one or more envelope segments depending on which key on the keyboard is being played. Envelope tracking is most often used to give the higher notes shorter envelopes and the lower notes longer envelopes, mimicking the response characteristics of percussion-activated acoustic instruments, such as guitar and marimba.
event editing: An operation in a sequencer in which one musical event at a time is altered.
F
FFT: Fast Fourier transform. A quick method of performing a Fourier analysis on a sound. See Fourier analysis.
filter: (1) A device for eliminating selected frequencies from the sound spectrum of a signal and perhaps (in the case of a resonant filter) increasing the level of other frequencies. See lowpass filter. (2) A device (MIDI filter) that eliminates selected messages from the MIDI data stream.
FM: See frequency modulation. 
FM synthesis: A technique in which frequency modulation (FM) is used to create complex audio waveforms. See frequency modulation.
formant: A resonant peak in a frequency spectrum. For example, the variable formants produced by the human vocal tract are what give vowels their characteristic sound.
Fourier analysis: A technique, usually performed using a DSP algorithm, that allows complex, dynamically changing audio waveforms to be described mathematically as sums of sine waves at various frequencies and amplitudes. See DSP.
frame: The basic unit of SMPTE time code, corresponding to one frame of a film or video image. Depending on the format used, SMPTE time can be defined with 24, 25, 30, or 29.97 frames per second. See SMPTE time code.
FreeMIDI: A Macintosh operating system extension developed by Mark of the Unicorn that enables different programs to share MIDI data. For example, a sequencer could communicate with a librarian program to display synthesizer patch names -- rather than just numbers -- in the sequencer's editing windows.
frequency modulation (FM): A change in the frequency (pitch) of a signal. At low modulation rates, FM is perceived as vibrato or some type of trill, depending on the shape of the modulating waveform. When the modulating wave is in the audio range (above 20Hz or so), FM is perceived as a change in tone color. FM synthesizers, commonly found on computer soundcards, create sounds using audio-range frequency modulation.
G
gain: The amount of boost or attenuation of a signal.
General MIDI (GM): A set of requirements for MIDI devices aimed at ensuring consistent playback performance on all instruments bearing the GM logo. Some of the requirements include 24-voice polyphony and a standardized group (and location) of sounds. For example, patch #17 will always be a drawbar organ sound on all General MIDI instruments. 
glide: A function, also called portamento, in which the pitch slides smoothly from one note to the next instead of jumping over the intervening pitches.
gigabyte: One billion (for British viewers, one thousand million) bytes.
global: Pertaining to or governing all of the operations of an instrument.
graphic editing: A method of editing parameter values using graphic representations (for example, of envelope shapes) displayed on a computer screen or LCD.
H
hard disk recording: A computer-based form of tapeless recording in which incoming audio is converted into digital data and stored on a hard disk.
harmonic: A frequency that is a whole-number multiple of the fundamental frequency. For example, if the fundamental frequency of a sound is 440Hz, then the first two harmonics are 880Hz and 1,320Hz (1.32kHz). See overtone.
headroom: The amount of additional signal above the nominal input level that can be sent into or out of an electronic device before clipping distortion occurs.
Hertz (Hz): the unit measurement of frequency. One Hz equals one cycle per second. The frequency range of human hearing is from 20Hz to 20kHz (20,000Hz).
highpass filter: A filter that attenuates the frequencies below its cutoff frequency.
Hz: See Hertz. 
I
inharmonic: Containing frequencies that are not whole-number multiples of the fundamental. See harmonic.
interface: A linkage between two things. A user interface is the system of controls with which the user controls a device. Two devices are said to be interfaced when their operations are linked electronically. An interface box is often required to convert signals from one form to another. For example, in order to get MIDI data in and out of a computer, you need some type of MIDI interface hardware. This may hook to an existing port on the computer, such as the printer port, or (in the case of the IBM-PC) it may consist of a circuit board that is plugged into one of the computer's internal slots. 
IRQ: Interrupt Request level. In IBM-PCs, a setting given to peripheral devices like soundcards and CD-ROM drives that identifies them to the computer's CPU. When the peripheral needs to communicate with the CPU, it will send an interrupt with that value. Problems will result if two or more peripherals are set to the same IRQ value.
J
K
keyboard scaling: A function with which the sound can be altered smoothly across the range of the keyboard by using key number as a modulation source. Level scaling changes the loudness of the sound, while filter scaling changes its brightness.
key follow: See envelope tracking.
keyboard tracking: See envelope tracking.
kHz: kilohertz (thousands of Hertz). See Hertz.
kilobyte (Kb): Linguistically speaking, a thousand bytes. In practice, a kilobyte generally contains 1,024 bytes.
L
layering: Sounding two or more voices, each of which typically has its own timbre, from each key depression. Layering can be accomplished within a single synthesizer, or by linking two synths together via MIDI and assigning both to the same MIDI channel.
LFO: Low-frequency oscillator. An oscillator especially devoted to applications below the audible frequency range, and typically used as a control source for modulating a sound to create vibrato, tremolo, trills, and so on.
librarian: See editor/librarian.
loop: A piece of material that plays over and over. In a sequencer, a loop repeats a musical phrase. In a sampler, loops are used to allow samples of finite length to be sustained indefinitely.
lowpass filter: A filter that attenuates (reduces in level) the frequencies above its cutoff frequency
M
map: A table in which input values are assigned to outputs arbitrarily by the user on an item-by-item basis. 
mapper: A device that translates MIDI data from one form to another in real time.
matrix modulation: A method of connecting modulation sources to destinations in such a way that any source can be sent to any combination of destinations.
Mb: See megabyte.
MCI: Media control interface. A multimedia specification designed to provide control of onscreen movies and peripherals like CD-ROM drives.
megabyte (Mb): Linguistically speaking, a million bytes. In practice, a megabyte often contains 1,024 kilobytes.
memory: A system or device for storing information -- in the case of musical devices, information about patches, sequences, waveforms, and so on.
merger: A MIDI accessory that allows two incoming MIDI signals to be combined into one MIDI output.
MIDI: (Musical Instrument Digital Interface) A communications standard for relaying information about a sound from an application or digital musical instrument to a synthesizer chip. MIDI commands contain all the information a sound board needs to reproduce the desired sound. MIDI is a specification for the types of control signals that can be sent from one electronic music device to another.
MIDI clock: A timing reference signal sent over a MIDI cable at the rate of 24 clock pulses per quarter-note (ppq).
MIDI Mapper: A Windows applet that automatically maps (shifts the value of) channel, program change, and note numbers. For example, a map could cause all notes coming in on MIDI channel 3 to go out on MIDI channel 7.
MIDI mode: Any of the ways of responding to incoming MIDI data. While four modes -- omni off/poly, omni on/poly, omni off/mono, and omni on/mono -- are defined by the MIDI specification, omni on/mono is never used, and at least two other useful modes have been developed -- multi mode for multitimbral instruments and multi-mono for guitar synthesizers.
MIDI Out/Thru: A MIDI output port that can be configured either to transmit MIDI messages generated within the unit (Out) or to retransmit messages received at the MIDI In (Thru).
MIDI thru: There are two types of MIDI thru. One, a simple hardware connection, is found on the back panels of many synthesizers. The thru jack in this case simply duplicates whatever data is arriving at the MIDI in jack. Sequencers have a second type, called software thru. In this case, data arriving at the in jack is merged with data being played by the sequencer, and both sets of data appear in a single stream at the out (not the thru) jack. A software thru is useful because it allows you to hook a master keyboard to the sequencer's MIDI input and a tone module to its output. You can then play the keyboard and hear the tone module, and the sequencer can also send its messages directly to the tone module. 
millisecond (ms): One one-thousandth (0.001) of a second. 
mixer: A device that adds two or more audio signals together.
mod: (1) Modulation. (2) Modification.
modem: A device (modulator/demodulator) that allows computer information to be sent over a telephone line.
modulation: The process of sending a control signal to a sound source so as to change the character of the sound.
module: A hardware sound generator with no attached keyboard. A module can be either physically separate or integrated into a modular synthesizer, and is designed to make some particular contribution to the process of generating electronic sound.
mod wheel: A controller, normally mounted at the left end of the keyboard and played with the left hand, that is used for modulation. It is typically set up to add vibrato. See modulation, vibrato.
mono mode: One of the basic reception modes of MIDI devices. In mono mode, an instrument responds monophonically to all notes arriving over a specific MIDI channel.
monophonic: Capable of producing only one note at a time.
MPC: Multimedia Personal Computer. A specification stating the minimum hardware requirements a computer must meet to display the MPC logo. They include 2Mb of RAM, a 16MHz 386SX processor, and 8-bit sound capabilities. This specification was published in 1990, and has since been bettered by the MPC 2 spec.
MPC 2: Multimedia PC, level 2. This specification requires the same types of hardware as MPC level 1, but with increased power and capacity. For example, 4Mb of RAM, a 25MHz 486SX processor, and 16-bit sound capabilities are specified.
ms: See millisecond.
MTC: MIDI time code. MTC is a way of transmitting SMPTE timing data over a MIDI cable. See SMPTE time code.
multi mode: A MIDI reception mode in which a multitimbral module responds to MIDI input on two or more channels and maintains musical independence between the channels, typically playing a different patch on each channel.
multisample: The distribution of several related samples at different pitches across the keyboard. Multisampling can provide greater realism in sample playback (wavetable) synthesis, since the individual samples don't have to be transposed over a great distance.
multitimbral: Capable of making more than one tone color (timbre) at the same time. A typical multitimbral tone generator can play, for example, the brass, piano, and violin parts all at once. 
N
normalize: To boost the level of a waveform to its maximum amount short of clipping (distortion). This maximizes resolution and minimizes certain types of noise.
Nyquist frequency: The highest frequency that can be reproduced accurately when a signal is digitally encoded at a given sample rate. Theoretically, the Nyquist frequency is half of the sampling rate. For example, when a digital recording uses a sampling rate of 44.1kHz, the Nyquist frequency is 22.050kHz. If a signal being sampled contains frequency components that are above the Nyquist limit, aliasing will be introduced in the digital representation of the signal unless those frequencies are filtered out prior to digital encoding. See aliasing, brick-wall filter.
O
omni mode: A MIDI reception mode in which a module responds to incoming MIDI channel messages no matter what their channel.
OMS: Open Music System (formerly Opcode MIDI System). A real-time MIDI operating system for Macintosh applications (and slated to be integrated into Windows 95). OMS allows communication between different MIDI programs and hardware, so that, for example, a sequencer could interface with a librarian program to display synthesizer patch names -- rather than just numbers -- in the sequencer's editing windows.
operator: A term used in Yamaha's FM synthesizers to refer to the software equivalent of an oscillator, envelope generator, and envelope-controlled amplifier.
oscillator: An electronic sound source. In an analog synthesizer, oscillators typically produce regularly repeating fluctuations in voltage; that is, they oscillate. In a digital synth, an oscillator more typically plays back a complex waveform by reading the numbers in a wavetable.
overdub: To record additional parts alongside (or merged with) previous tracks. Overdubbing enables "one-man band" productions, as multiple synchronized performances are recorded sequentially.
overtone: A whole-number multiple of the fundamental frequency of a tone. The overtones define the harmonic spectrum of a sound. See Fourier analysis, partial.
P
parallel interface: A connection between two pieces of hardware in which several data lines carry information at the same time. Compare with serial interface.
parameter: A user-adjustable quantity that governs some aspect of a device's performance. Normally, the settings for all of the parameters that make up a synthesizer patch can be changed by the user and stored in memory, but the parameters themselves are defined by the operating system and cannot be altered.
partial: One of the sine-wave components (the fundamental, an overtone, or a tone at some other frequency) of a complex tone. See overtone.
patch: Verb: To connect together, as the inputs and outputs of various modules, generally with patch cords. Noun: The configuration of hookups and settings that results from the process of patching, and, by extension, the sound that such a configuration creates. Often used to denote a single tone color or the contents of a memory location that contains parameter settings for such a tone color, even on an instrument that requires no physical patching.
patch map: A map with which any incoming MIDI program change message can be assigned to call up any of an instrument's patches (sounds). See map, MIDI Mapper.
PCM: Pulse code modulation -- a standard method of encoding analog audio signals in digital form.
percentage quantization: A method of quantization in which notes recorded into a sequencer with uneven rhythms are not shifted all the way to their theoretically perfect timings but instead are shifted part of the way, with the amount of shift being dependent on the user-selected percentage (quantization strength). See quantization.
physical modeling synthesis: A type of sound synthesis performed by computer models of instruments.This technique emulates the impulse patterns of real-world instruments using a software
model. These models are sets of complex equations that describe the physical properties of an instrument (such as the shape of the bell and the density of the material) and the way a musician interacts with it (blow, pluck, or hit, for example). 
pitch-bend: A shift in a note's pitch, usually in small increments, caused by the movement of a pitch-bend wheel or lever; also, the MIDI data used to create such a shift. See bend.
pitch-shift: To change the pitch of a sound without changing its duration, as opposed to pitch-transpose, which changes both. Some people use the two terms interchangeably.
plug-in: A software program that acts as an extension to a larger program, adding new features. 
pole: A portion of a filter circuit. The more poles a filter has, the more abrupt its cutoff slope will be. Each pole causes a slope of 6dB per octave; typical filter configurations are two-pole (12dB/oct) and four-pole (24dB/oct). See rolloff slope.
poly mode: A MIDI reception mode in which a module responds to note messages on only one channel, and plays as many of these notes at a time (polyphonically) as it can.
polyphonic: Capable of producing more than one note at a time. All synthesizers place a limit on how many voices of polyphony are available. General MIDI-compliant synthesizers are required to provide 24 voices of polyphony. Compare with multitimbral.
polyphony: The number of voices (notes) a device can produce simultaneously.
poly pressure: Polyphonic pressure. (Also called key pressure.) A type of MIDI channel message in which each key senses and transmits pressure data independently. Compare with channel pressure.
port: Verb: To translate a program written for one computer so that it can be 
run on a different model. Noun: An electrical connector of some specialized type, e.g., SCSI port, MIDI port, serial port.
portamento: See glide.
pot: Potentiometer. A device (commonly attached to a knob or slider) used to adjust some aspect of the signal being passed through it, or to send out a control signal corresponding to its position.
ppq: Pulses per quarter-note; the usual measure of a sequencer's clock resolution.
preset: (1) A factory-programmed patch that cannot be altered by the user. (2) Any patch. Note: Some manufacturers make distinctions between presets, programs, and/or patches, each of which may contain a different set of parameters.
pressure sensitivity: See aftertouch, channel pressure, poly pressure.
program: Verb: To create a synthesizer patch. Noun: A patch. See patch, preset.
program change: A MIDI message that causes a synthesizer or other device to switch to a new program (also called preset, patch) contained in its memory.
programmable: Equipped with software that enables the user to create new sounds or other assignments by altering parameter settings and storing the new settings in memory. An individual control parameter is said to be programmable if its setting can be stored separately with each individual patch.
Q
quantization: A function found on sequencers and drum machines that causes notes played at odd times to be "rounded off" to regular rhythmic values. See percentage quantization.
quantization noise: One of the types of error introduced into an analog audio signal by encoding it in digital form. The digital equivalent of tape hiss, quantization noise is caused by the small differences between the actual amplitudes of the points being sampled and the bit resolution of the analog-to-digital converter.
quantized: Set up to produce an output in discrete steps.
QuickTime: A software multimedia environment developed by Apple Computer, running on the Macintosh or under Windows 3.1. QuickTime enables the creation and playback of QuickTime movies featuring full-motion video, MIDI tracks and 16-bit ADPCM audio. 
R
RAM: Random access memory. RAM is used for storing user-programmed patch parameter settings in synthesizers, and sample waveforms in samplers. A constant source of power (usually a long-lasting battery) is required for RAM to maintain its contents when power is switched off. Compare with ROM.
real time: Occurring at the same time as other, usually human, activities. In real-time sequence recording, timing information is encoded along with the note data by analyzing the timing of the input. In real-time editing, changes in parameter settings can be heard immediately, without the need to play a new note or wait for computational processes to be completed.
reconstruction filter: A lowpass filter on the output of a digital-to-analog converter that smoothes the staircase-like changes in voltage produced by the converter in order to eliminate clock noise from the output.
release: The portion of an envelope that begins after the key is lifted. See ADSR.
release velocity: The speed with which a key is raised, and the type of MIDI data used to encode that speed. Release velocity sensing is rare but found on some instruments. It is usually used to control the rate of the release segments of the envelope(s).
resolution: The fineness of the divisions into which a sensing or encoding system is divided. The higher the resolution, the more accurate the digital representation of the original signal will be.
resonance: A function on a filter in which a narrow band of frequencies (the resonant peak) becomes relatively more prominent. If the resonant peak is high enough, the filter will begin to oscillate, producing an audio output even in the absence of input. Filter resonance is also known as emphasis and Q. It is also referred to in some older instruments as regeneration or feedback, because feedback was used in the circuit to produce a resonant peak.
reverb: A type of digital signal processing that produces a continuous wash of echoing sound, simulating an acoustic space such as a concert hall. Reverberation contains the some frequency components as the sound being processed, but no discrete echoes. See echo, DSP.
ring modulator: A special type of mixer that accepts two signals as audio inputs and produces their sum and difference tones at its output, but does not pass on the frequencies found in the original signals themselves. See clangorous.
rolloff slope: The acuity of a filter's cutoff frequency. Rolloff is generally measured in decibels (dB) per octave. A shallow slope, such as 6dB per octave, allows some frequency components beyond the cutoff frequency to be heard, but at a reduced volume. When the rolloff slope is steep (on the order of 24dB per octave), frequency components very close to the cutoff frequency are reduced in volume so much that they fall below the threshold of audibility. See filter, pole.
ROM: Read-only memory. A type of data storage whose contents cannot be altered by the user. An instrument's operating system, and in some cases its waveforms and factory presets, are stored in ROM. Compare with RAM.
S
sample: Noun: A digitally recorded representation of a sound. Also, a single word of the data that makes up such a recording. See word. Verb: To make a digital recording. See sampling.
sample-and-hold: A circuit on an analog synthesizer that, when triggered (usually by a clock pulse), looks at (samples) the voltage at its input and then passes this voltage on to its output unchanged, regardless of what the input voltage does in the meantime (the hold period), until the next trigger is received. In one familiar application, the input was a noise source and the output was connected to oscillator pitch, which caused the pitch to change in a random staircase pattern. The sample-and-hold effect is often emulated by digital synthesizers through an LFO waveshape called "random."
sampler: An instrument that records and plays back samples, usually by allowing them to be distributed across a keyboard and played back at various pitches.
sampling: The process of encoding an analog signal in digital form by reading (sampling) its level at precisely spaced intervals of time. See sample, sampling rate.
sampling rate: The number of samples taken per second. Typical sampling rates vary from 11kHz to 48kHz. See sampling, Nyquist frequency.
scrub: To move backward and forward through an audio waveform under manual control, in order to find a precise point in the wave for editing purposes.
SCSI: Small Computer Systems Interface, a high-speed communications protocol that allows computers, samplers, and disk drives to communicate with one another. Pronounced "scuzzy."
SDII: Sound Designer II, an audio file format. The native format of Digidesign's Sound Designer II (Macintosh) graphic audio waveform editing program.
SDS: The MIDI sample dump standard. SDS is used to transfer digital audio samples from one instrument to another over a MIDI cable.
sequence: A set of music performance commands (notes and controller data) stored in a sequencer.
sequencer: A device or program that records and plays back user-determined sets of music performance commands, usually in the form of MIDI data. Most sequencers also allow the data to be edited in various ways, and stored on disk.
serial interface: An electronic connection between two devices in which digital data is transferred one bit after another, rather than several bits at a time. MIDI is a serial interface. Compare with parallel interface.
SFI: A file extension specifying Turtle Beach's SoundStage audio format. Typically encountered as FILENAME.SFI.
sidebands: Frequency components outside the natural harmonic series, generally introduced to the tone by using an audio-range wave for modulation. See clangorous.
single-step mode: A method of loading events (such as notes) into memory one event at a time. Also called step mode and step-time. Compare with real time.
sine wave: A signal put out by an oscillator in which the voltage or equivalent rises and falls smoothly and symmetrically, following the trigonometric formula for the sine function. Sub-audio sine waves are used to modulate other waveforms to produce vibrato and tremolo. Audio-range sine waves contain only the fundamental frequency, with no overtones, and thus can form the building blocks for more complex sounds.
SMDI: SCSI musical data interchange. A specification for sending MIDI sample dumps over the SCSI bus. See SDS.
SMP: Turtle Beach's SampleVision audio file format. Typically encountered as FILENAME.SMP.
SMPTE time code: A timing reference signal developed by the Society of Motion Picture & Television Engineers and used for synchronizing film and videotape to audio tape and software-based playback systems. Pronounced "simp-tee." See frame.
SND: Sound resource. A Macintosh audio file format.
snapshot automation: A form of mixing automation (frequently MIDI-controlled) in which the controlling device records the instantaneous settings (the snapshot) for all levels and pan pots, and recalls these settings on cue.
song position pointer (SPP): A type of MIDI data that tells a device how many sixteenth-notes have passed since the beginning of a song. An SPP message is generally sent in conjunction with a continue message in order to start playback from the middle of a song.
sostenuto pedal: A pedal found on the grand piano and mimicked on some synthesizers, with which notes are sustained only if they are already being held on the keyboard at the moment when the pedal is pressed. Compare with sustain pedal.
soundcard: A circuit board that installs inside a computer (typically an IBM-compatible) adding new sound capabilities. These capabilities can include an FM or wavetable synthesizer and audio inputs and outputs. MIDI inputs and outputs are also normally included.
split keyboard: A single keyboard divided electronically to act as if it were two or more separate ones. The output of each note range is routed into a separate signal path in the keyboard's internal sound-producing circuitry, or transmitted over one or more separate MIDI channels. Applications include playing a bass sound with the left hand while playing a piano sound with the right.
SPP: See song position pointer.
status byte: A MIDI byte that defines the meaning of the data bytes that follow it. MIDI status bytes always begin with a 1 (hex 8 through F), while data bytes always begin with a 0 (hex 0 through 7).
step input: In sequencing, a technique that allows you to enter notes one step at a time. (Also called step recording.) Common step values are sixteenth- and eighth-notes. After each entry, the sequencer's clock (position in the sequence) will advance one step, then stop, awaiting new input. Recording while the clock is running is called real-time input. 
subtractive synthesis: The technique of arriving at a desired tone color by filtering waveforms rich in harmonics. Subtractive synthesis is the type generally used on analog synthesizers. Compare with FM synthesis, sampling.
sustain: The third of the four segments in an ADSR envelope. The sustain portion of the envelope begins when the attack and decay portions have run their course, and continues until the key is released. The sustain control is used to determine the level at which the envelope will remain. While the attack, decay, and release controls are rate or time controls, the sustain control is a level control.
sustain pedal: The electronic equivalent of a piano's damper pedal. In most synthesizers, the sustain pedal latches the envelopes of any currently playing or subsequently played notes at their sustain levels, even if the keys are lifted. 
sync: Synchronization. Two devices are said to be in sync when they are locked together with respect to time, so that the events generated by each of them will always fall into predicable time relationships.
sync track: A timing reference signal recorded onto tape. See SMPTE time code, FSK.
synthesizer: A musical instrument that generates sound electronically and is designed according to certain principles developed by Robert Moog and others in the 1960s. A synthesizer is distinguished from an electronic piano or electronic organ by the fact that its sounds can be programmed by the user, and from a sampler by the fact that the sampler allows the user to make digital recordings of external sound sources.
system-common: A type of MIDI data used to control certain aspects of the operation of the entire MIDI system. System-common messages include song position pointer, song select, tune request, and end-of-system-exclusive.
system-exclusive (sys-ex): A type of MIDI data that allows messages to be sent over a MIDI cable that will be responded to only by devices of a specific type. Sys-ex data is used most commonly for sending patch parameter data to and from an editor/librarian program.
system real-time: A type of MIDI data that is used for timing reference. Because of its timing-critical nature, a system real-time byte can be inserted into the middle of any multi-byte MIDI message. System real-time messages include MIDI clock, start, stop, continue, active sensing, and system reset.
T
THD: Total harmonic distortion. An audio measurement specification used to determine the accuracy with which a device can reproduce an input signal at its output. THD describes the cumulative level of the harmonic overtones that the device being tested adds to an input sine wave. THD+n is a specification that includes both harmonic distortion of the sine wave and nonharmonic noise.
timbre: (1) Tone color. (2) One of the building blocks of a patch in a Roland synthesizer. Pronounced "tam-br."
time code: A type of signal that contains information about location in time. Used for a synchronization reference when synchronizing two or more machines such as sequencers, drum machines, and tape decks.
touch-sensitive: Equipped with a sensing mechanism that responds to variations in key velocity or pressure by sending out a corresponding control signal. See velocity, aftertouch.
track: Verb: To be controlled by or follow in some proportional relationship (as when a filter's cutoff frequency tracks the keyboard, moving up or down depending on what note is played). Noun: One of a number of independent memory areas in o sequencer. By analogy with tape tracks, sequencer tracks are normally longitudinal with respect to time and play back in sync with other tracks. 
transient: Any of the non-sustaining, non-periodic frequency components of a sound, usually of brief duration and higher amplitude than the sustaining components, and occurring near the onset of the sound (attack transients).
tremolo: A periodic change in amplitude, usually controlled by an LFO, with a periodicity of less than 20Hz. Compare with vibrato.
U
upload: to transfer a file from a computer to an electronic bulletin board (BBS), usually via modem. See download.
V
VCA: Voltage-controlled amplifier. A device that responds to a change in voltage at its control input by altering the gain of a signal being passed through it. Also, the digital equivalent of a VCA.
VCF: Voltage-controlled filter. A filter whose cutoff frequency can be changed by altering the amount of voltage being sent to its control input. Also, the digital equivalent of a VCF.
VCO: Voltage-controlled oscillator. An oscillator whose frequency can be changed by altering the amount of voltage being sent to its control input.
velocity: A type of MIDI data (range 1 to 127) usually used to indicate how quickly a key was pushed down (attack velocity) or allowed to rise (release velocity). Note: A note-on message with a velocity value of 0 is equivalent to a note-off message.
velocity curve: A map that translates incoming velocity values into other velocities in order to alter the feel or response of a keyboard or tone module.
velocity sensitivity: A type of touch sensitivity in which the keyboard measures how fast each key is descending. Compare with pressure sensitivity.
vibrato: A periodic change in frequency, often controlled by an LFO, with a periodicity of less than 20Hz. Compare with tremolo.
virtual: Existing only in software.
VOC: A file extension specifying the Creative Labs Sound Blaster audio format. Typically encountered as FILENAME.VOC.
voice: (1) An element of synthesizer circuitry capable of producing a note. The polyphonic capability of a synthesizer is defined by how many voices it has. See polyphony. (2) In Yamaha synthesizers, a patch (sound).
voice channel: A signal path containing (at a minimum) an oscillator and VCA or their digital equivalent, and capable of producing a note. On a typical synthesizer, two or more voice channels, each with its own waveform and parameter settings, can be combined to form a single note.
voice stealing: A process in which a synthesizer that is being required to play more notes than it has available voices switches off some currently sounding voices (typically those that have been sounding longest or are at the lowest amplitude) in order to assign them to play new notes.
W
.WAV: The Windows audio file format. Typically encountered as FILENAME.WAV.
waveform: A signal, either sampled (digitally recorded) or periodic, being generated by an oscillator. Also, the graphic representation of this signal, as on a computer screen. Each waveform has its own unique harmonic content. See oscillator.
waveshape: See waveform.
wavetable synthesis: A common method for generating sound electronically on a synthesizer or PC. Output is produced using a table of sound samples--actual recorded sounds--that are digitized and played back as needed. By continuously rereading samples and looping them together at different pitches, highly complex tones can be generated from a minimum of stored data without overtaxing the processor.
wavetable lookup: The process of reading the numbers in a wavetable (not necessarily in linear order from beginning to end) and sending them to a voice channel.
wet: Consisting entirely of processed sound. The output of an effects device is 100% wet when only the output of the processor itself is being heard, with none of the dry (unprocessed) signal. Compare with dry.
wheel: A controller, normally mounted at the left end of the keyboard and played with the left hand, that is used for pitch-bending or modulation.
word: A single number (sample word) that represents the instantaneous amplitude of a sampled sound at a particular moment in time. In 8-bit recording, a sample word contains one byte; in 16-bit recording, each word is a two-byte number.
workstation: A synthesizer or sampler in which several of the tasks usually associated with electronic music production, such as sequencing, effects processing, rhythm programming, and data storage on disk, can all be performed by components found within a single physical device.
X
Y
Z
zero crossing: A point at which a digitally encoded waveform crosses the center of its amplitude range. 
zone: A contiguous set of keys on the keyboard. Typically, a single sound or MIDI channel is assigned to a given zone.