AP - The Recognized Standard In Audio Test

Audio Terms Glossary HOME/LIBRARY/GLOSSARY

This list of technical terms is pulled from the Audio Measurement Handbook.

 

A
A/D converter	see analog-to-digital converter.
ac mains	the utility-provided ac power source from which most electronics equipment is operated. Also called power line, ac line and line voltage.
AC-2	A low-bit-rate transmission technique developed for broadcast and other professional applications by Dolby Laboratories. AC-2 is a transform-based technology.
acceptance limit	In pass/fail testing, the poorest measured performance specification which a manufacturer is willing to accept and ship in a product.
accuracy	the degree of conformance of a test instrument to absolute standards, usually expressed as a percentage of reading or a percentage of measurement range (full scale).
acoustical testing	audio-frequency tests in which the signal exists as acoustical waves which must be transformed into electrical signals by a transducer such as a microphone before measurement.
ADC	see analog-to-digital converter.
AES	the Audio Engineering Society, with headquarters in New York City.
AES/EBU interface	See digital audio signal, bi-phase coded.
AES17-1998	a standard promulgated by the AES providing methods for specifying and verifying the performance of digital audio equipment.
AES3 interface	a digital interface standard for professional audio equipment interconnection. Also AES/EBU, IEC60958. See digital audio signal, bi-phase coded.
AGC	see automatic gain control.
alias	a nonlinear signal product falling within the audio band, caused by the presence of an out-of-band (above 1/2 the sampling rate) signal during the original A-to-D conversion. Alias signals fall at sum and difference frequencies of the original signal frequency (or its harmonics) and the sampling rate (or its harmonics).
amplitude	the magnitude of a signal, which may be expressed in a wide variety of units such as volts, dBm, dBu, watts, etc. Also level.
analog recording, processing	techniques in which a signal is represented, recorded, processed, or transmitted as a continuously variable quantity.
analog-to-digital converter	a device for converting an analog input signal into a series of digital values representing the instantaneous amplitude of the signal at regular sampling intervals. Abbreviated ADC. See digital recording or processing.
anechoic chamber	a room, or space treated with sound-absorbent material so as to have no significant acoustical reflections.
ANSI	American National Standards Institute, a U.S.-based standards organization.
anti-alias filter	a low-pass filter preceding an ADC to prevent signals at frequencies greater than one-half the sampling rate from reaching the converter. Signals above one-half the sampling rate cannot be unambiguously converted and would appear in the digital output as signals at incorrect frequencies.
AP2700	the Audio Precision control software for the 2700 series of audio test and measurement instruments.
APT-X™	a low-bit-rate coding system for professional applications, manufactured by Audio Processing Technology. It divides the audio range into four bands and digitally compresses each.
APWIN	the Audio Precision control software for the System One, System Two and System Two Cascade lines of audio test and measurement instruments.
APx500	the Audio Precision measurement software for the APx Series 500 analog analyzers.
ASCII	American Standard Code for Information Interchange, a standard for representing 127 alphabetic, numeric, and other characters and symbols as 7-bit binary numbers.
ATRAC	Adaptive Transform Acoustic Coding, the perceptually-based low-bit-rate coding technique used in the Sony MiniDisc system.
ATS	the Audio Precision control software for ATS-2 audio test and measurement instruments.
Attack time	the time interval required following a sudden increase in input level to a device for the output of the device to reach some stated percentage of the eventual signal level.
attenuator	a passive network of resistors (plus possibly compensating capacitors) which reduces the amplitude of a signal by a precise amount. Often also referred to as a pad.
automatic gain control	a technique for automatically varying the gain of a device, usually so as to maintain a relatively constant output level. Abbreviated AGC.
automatic volume control	a technique for maintaining output audio level (volume) relatively constant in spite of varying input levels. Abbreviated AVC.
AVC	see automatic volume control.
average responding	a form of ac signal detection indicative of the average absolute value of a waveform, typical of analog meter movements.
A-weighting filter	a specific noise-weighting filter (ANSI S1.4, IEC Recommendation 179) used to produce noise measurements which correlate well with human observations. See weighting filter.
B
back termination	a specified value of resistance or impedance connected for test purposes to a device input instead of the normal source which drives the device under test.
balance	the degree of matching of the two sides of a balanced input or output configuration, or of the levels on the two channels of a stereo system.
balanced	a term referring to an audio transmission line in which the signal is applied differentially between two conductors, each of which has equal impedance to a common reference or ground. A balanced line is usually constructed with three conductors: an internal twisted pair of wires, one carrying signal “+” or HIGH, the other carrying signal “–” or LOW. These are surrounded by a third conductor in the form of a braided or foil shield, which is connected to the common or ground terminal at one or both ends of the cable. Used with properly engineered equipment, balanced lines are superior in performance to unbalanced connections, yielding better rejection of common-mode interference caused by electrostatic and electromagnetic fields. Additionally, since the audio in a balanced circuit is isolated from the ground conductor, ground-current-induced noise is much more easily dealt with. Also called a symmetrical line. See ground, unbalanced.
bandpass filter	a filter that passes a specific frequency band (called the passband) essentially without attenuation while attenuating frequencies both below and above the specified band.
band-reject filter	a filter which attenuates the frequencies within a specified frequency range while passing essentially without attenuation the frequencies below and above that range. If the rejection range is narrow, a band reject filter is also often called a notch filter.
Bessel filters	a family of filters possessing maximally flat time delay as a function of frequency. Bessel filters are thus a time domain optimization rather than a frequency domain optimization.
Bessel function	a mathematical function which describes the amplitudes of the carrier and sidebands resulting from a frequency modulation process.
bias	in analog tape recording, a high-frequency ac signal (typically at least 60 kHz, often much higher) combined with the audio signal in the recording head of an analog tape recorder. Proper bias level reduces the distortion of the magnetic recording process. A signal derived from the same oscillator is normally also used in the erase head to remove previous recordings from the tape.
bin, FFT	the basic frequency resolution-determining division of an FFT-computed spectrum. The FFT process cannot resolve signals of different frequency falling into the same bin. The bin width of an FFT can be computed from (sample rate)/(number of amplitude samples) in the digitized waveform used as input to the FFT process. The terms bin and line are used interchangeably.
bit	binary digit, which may have only two possible states (ON/OFF, HIGH/LOW, 1/0, etc.)
bit depth	see word length.
bits of resolution	see word length.
Bode plot	a graph of the input-output gain and phase relationships versus frequency of an amplifier.
bridging	a relatively high impedance input (usually balanced) which may be connected across a lower impedance circuit without significantly affecting the levels in the lower impedance circuit.
BTSC	Broadcast Television Systems Committee of the Electronic Industries Association, the group which developed the standards for the stereo audio transmission techniques used in North American television broadcasting. The technique itself is commonly referred to as BTSC stereo.
bus	in electricity and electronics, a conductor common to three or more circuits, such as a ground bus. In audio mixing consoles, a circuit or system that sums two or more inputs, such as a mixing bus or a send bus. In computers, the data bus is a set of conductors that carry common data between three or more subunits of the computer.
buss	a kiss; only tangentially related to electricity.
Butterworth filters	a family of filters possessing maximally-flat amplitude response as a function of frequency with no ripples in the passband. Butterworth filters can be high-pass or low-pass.
byte	an 8-bit digital word.
C
CCIR	Comite Consultatif International des Radiocommunications (International Radio Consultative Committee), an international standards-setting organization with headquarters in Geneva, Switzerland.
CCIR 468	a CCIR specification which includes, among other things, standards for weighted and unweighted noise measurements. The weighted standard specifies the CCIR weighting filter and a quasi-peak detector (see weighting filter). The unweighted standard specifies a 22 Hz to 22 kHz bandwidth limiting filter and an rms detector.
CCIR-ARM	a noise measurement technique developed by Dolby Laboratories, which uses the weighting filter shape specified in CCIR Recommendation 468 but with the unity-gain frequency at 2 kHz rather than 1 kHz, and an average-responding detector. Also CCIR-2k.
CCIR-2k	see CCIR-ARM
CCITT	Comite Consultatif International Telephonique et Telegraphique (International Telegraph and Telephone Consultative Committee), an international standards organization for telephone communications related industries.
CCITT O.33	a standard, adopted from the EBU R-27 recommendation, for testing broadcast transmission links. The O.33 standard starts with an audio frequency-shift-keyed header which identifies the type of test sequence to follow and the source of the signal. The remainder of the test sequence is a series of tones of specified level and frequency, plus a quiet period for noise measurements. A compatible analyzer at the other end of the link is synchronized by the FSK header and then makes measurements from each section of the signal sequence. Each tone in the sequence is normally of one second in duration.
CCITT weighting filter	a noise-weighting filter described in CCITT documents P53A and P53B. See weighting filter.
CD	compact disc, a digital recording system based on optically-encoded discs and linear 16-bit coding at a 44.1 kHz sample rate.
Chebyshev filters	a family of filters with a sharper rolloff than Butterworth and possessing passband ripple of specific amplitude.
clipping	the action of a system in flattening and squaring off signal peaks when driven with a signal whose peak amplitude is beyond its linear signal-handling capability.
C-message weighting filter	a weighting filter commonly used for noise measurements in the telephone industry. The shape of the C-message weighting filter is based on both the typical human ear frequency response and also on the response of a typical telephone receiver. The C-message weighting filter is described in the IEEE-743 and Bell 41009 specifications. See weighting filter.
CMRR	see common mode rejection ratio.
CODEC	coder-decoder for reduced bit rate transmission or recording/reproduction of digital audio. Codecs reduce the required bit rate by using the available bit rate resources for the most important portions of the signal. More complex coders analyze the signal and allocate the bits in a more optimal form from moment to moment. Some coders are perceptual coders, in which signals which will be psychoacoustically masked because they are nearby in frequency to a stronger signal will be encoded with fewer bits or not at all.
common mode	signals on a balanced pair which are in phase with respect to ground, as opposed to differential signals which are developed across the balanced pair (normal mode). Common mode signals are sometimes also referred to as longitudinal signals.
common mode rejection ratio	the ability of a balanced (differential) input circuit to reject signals on the two conductors which are in phase with respect to common or ground.
common mode voltage range	the maximum voltage between either side of a balanced pair and ground or common which can be tolerated while still permitting full common mode rejection.
compander	a signal processing amplifier which can compress or expand (or both) the dynamic range of the output signal with respect to that of the input signal.
compressor	a signal processing amplifier whose gain automatically reduces with higher amplitude input signals, such that the dynamic signal range at its output is less than the signal dynamic range at its input.
console	see mixer.
consumer	In audio, often used to designate digital audio format. See digital audio signal, bi-phase coded.
continuous sweep	Audio Precision term for a proprietary log chirp measurement technique.
control software	for Audio Precision products, one of several software programs that control Audio Precision audio test and measurement instruments. APWIN, ATS and AP2700 are examples of Audio Precision control software.
crest factor	the ratio of a signal's peak amplitude to its rms amplitude.
critical band	in pyschacoustics, the maximum bandwidth of noise which is perceived by humans to be the same loudness as a sine wave of the same power at band center.
crossover distortion	a characteristic type of distortion produced in an amplifier’s push-pull output stage if improperly biased such that only the peaks of low-level signals drive the amplifier into normal amplification ranges. A “dead band” input amplitude range may consequently exist, with signals in the “dead band” not producing output.
crossover, loudspeaker	a filter network designed to separate the audio frequency spectrum into two or more ranges. This permits feeding the several drivers (woofer, tweeter, etc.) of a multi-driver loudspeaker system with only the frequency range of the signal for which they have been optimized.
crosstalk	unwanted signal coupling from one channel of a multi-channel transmission or recording system to another.
D
D/A converter	see digital-to-analog converter.
DAC	see digital-to-analog converter.
damping factor	the ratio of the rated load impedance of an amplifier to the output (source) impedance of the amplifier.
DASH	Digital Audio Stationary Head, a standard for the recording of digital audio signals onto magnetic tape. DASH equipment is now rare.
DAT	see RDAT.
dB	abbreviation for decibel.
dB/octave	decibels per octave. A standard means of referring to the ultimate rejection slope (attenuation versus frequency) of a band-limiting filter. Each pole of a band-limiting filter produces an ultimate rejection slope of 6 dB/octave (20 dB/decade). Thus, a 3-pole filter will have a rejection slope of 18 dB/octave (60 dB/decade).
dBFS	decibels referenced to digital full scale (FS), where 0 dBFS is the rms value of a sine wave whose positive peak just reaches positive full scale. For everyday use, 0 dBFS can be considered to be the maximum digital amplitude available within a system. However, it should be noted that because 0 dBFS has been defined for a sine wave, waveforms with lower crest factors can exceed 0 dBFS by as much as 3.01 dB before incurring digital clipping. See crest factor and digital clipping.
dBm	decibels relative to a reference value of 1 milliwatt. dBm is a power unit and requires knowledge of power levels (voltage and current, or voltage and impedance, or current and impedance) rather than merely voltage. This term has historically been associated with a measurement of audio levels in professional use, but it is rarely correctly applied and should not be used except in certain very specific circumstances. As a unit of power, the voltage value of a dBm will vary with the circuit impedance. Use the term dBu instead. See dBu.
dBr	decibels relative to an arbitrary reference value, r. The reference value must be stated for this to be a meaningful unit. The dBr can be a handy shortcut during tests, allowing you, for example, to measure a specific voltage and note it as your reference, 0 dBr; and then to express further measurements in dBr relative to this level.
dBspl	dB sound pressure level, a standard unit in acoustical measurements. 0 dBspl corresponds to a sound pressure of 20 micropascal (20 micronewtons per meter squared, 0.0002 dynes per square centimeter). Thus, +94 dBspl corresponds to one Pascal.
dBu	decibels relative to a signal level of 0.7746 V rms. dBu now is the common term for analog audio amplitude in professional audio interfaces and circuits. 0 dBu (0.7746 V rms) equals 0 dBm, but only in a 600 ohm load impedance. Unless you are clearly interested in measuring power in a circuit of known impedance, use dBu, not dBm. See dBm.
dBV	decibels relative to a signal level of 1 volt rms. dBV now is the common term for analog audio levels in consumer audio interfaces and circuits. 0 dBV equals +2.218 dBu.
dbx™	the family of noise reduction techniques developed by the dbx company.
DCC	Digital Compact Cassette, a low-bit-rate recording medium using the PASC perceptual coding technique, developed for consumer applications by Philips. DCC cassettes have the same physical dimensions as conventional analog cassettes, permitting design of machines which can play conventional cassettes in addition to recording and playing digital cassettes according to the DCC standard. Now rare.
decade	the interval between two frequencies with a ratio of exactly 10:1, such as the range from 20 Hz to 200 Hz, or from 1 kHz to 10 kHz.
decibel	decibel, a ratio unit for expressing signal amplitudes. If the amplitudes are expressions of voltage, dB = 20 log10 (V1/V2). If the amplitudes are expressions of power, dB = 10 log10 (P1/P2). . Two important points to remember: * The dB, mimicking our hearing, is not a linear unit but a measurement on a logarithmic scale. * Decibels are relative units of measure, having no meaning in an absolute sense. A dB must always be referenced to something to have meaning. You can speak of a “4 dB change” or a “60 dB signal-to-noise ratio” because these are both relative statements, but describe a signal level as 39 dB means nothing. All absolute measurements expressed as decibels must have an indication of reference, such as dBu, dBV, dBr and so on. See dBu, dBV, dBr.
decimation	a mathematical process of reducing the clock rate and bandwidth content of a digital audio signal.
deemphasis	signal processing, normally a high-frequency attenuation, to reduce the effects of noise generated in some previous sections of a transmission or recording system. In order to provide overall flat frequency response, preemphasis is performed prior to the noise-introducing section of the system. Common broadcasting deemphasis and preemphasis systems use single-pole filters and are referred to by the time constant of the resistor-capacitor network making up the circuit; for example, 50 µs or 75 µs deemphasis. See preemphasis.
detector	the precision ac to dc conversion section of a measurement instrument, located following all ac signal processing and prior to the indicating portion. Detectors are classified according to which parameter of the input ac signal the output dc value linearly follows-true rms, average, peak, etc.
difference product	an intermodulation distortion signal at the frequency which is the difference between two applied signal frequencies. Sometimes referred to as difference tone. For example, test signals of 13 kHz and 14 kHz will produce a 1 kHz difference product when applied to a device which has asymmetrical nonlinearity.
differential	balanced; a signal existing across two conductors, rather than between either conductor and ground or common.
digital audio signal, bi-phase coded	In the consumer and professional audio field, digital audio is typically carried from point to point as bi-phase coded signal, commonly referred to as AES3, AES/EBU or S/PDIF. There are electrical and bitstream protocol differences among the variations of bi-phase coded digital audio, but the various signals are largely compatible. See a, b and c below.
digital audio signal, bi-phase coded: a. protocols	The “professional” and “consumer” protocols are indicated in the digital signal’s status bits. With some exceptions, the protocols are compatible; even if status bit information is lost, the audio is usually passed. Although either protocol can be carried on any of the electrical or optical formats, systems using the professional formats typically use the professional protocol, and those using consumer formats use the consumer protocol. In APx500 series instruments, by default the digital audio protocol is set to “consumer.” When you are using the electrical digital output, you can optionally set the Professional checkbox in the Signal Path Setup view. This sets the electrical signal to the professional level, and also sets the protocol and status bits to “professional.” See b and c below.
digital audio signal, bi-phase coded: b. electrical formats	AES3, AES/EBU, IEC60958-4 all refer to an electrically balanced professional format, with a 2–7 volt pp signal in a 110 ohm impedance. AES-3id and SMPTE276M refer to an electrically unbalanced professional format, with a 1 volt pp signal in a 75 ohm impedance. S/PDIF and IEC60958-3 refer to an electrically unbalanced consumer format, with a 0.5 volt pp signal in high impedance. In APx500 series instruments, the digital audio output is unbalanced and is by default set to 0.5 volts pp, a “consumer” setting compatible with S/PDIF and IEC60958-3. If the Professional checkbox is set in the Signal Path Setup view, the level is set to 1 volt, compatible with AES-3id and SMPTE276M. The digital input is unbalanced and by default set to high impedance, compatible with S/PDIF and IEC60958-3. If the 75 Ohm Termination checkbox is set in the Signal Path Setup view, the impedance is set to 75 ohms, compatible with AES-3id and SMPTE276M. See a, above, and b, below.
digital audio signal, bi-phase coded: c. optical formats	Digital optical formats have been left to manufacturers to define. Ours is compatible with Toshiba’s Toslink format, as are most digital audio optical inputs and outputs. See a and b above.
digital recording or processing	a technique in which the original signal is periodically sampled and the amplitude value at each sampling instant is converted into a number represented by a binary word.
digital-to-analog converter	a device that converts a stream of digital numbers, each representing the amplitude of a signal at a particular sampling time, into a corresponding analog signal. Abbreviated DAC.
DIM	dynamic intermodulation distortion. A technique to measure nonlinearity of a device, designed to be particularly sensitive to distortions produced during transient conditions typical of program material. Also referred to as TIM—transient intermodulation distortion.
DIN	Deutsches Institute für Normung, a German standards organization.
DIN 45403	a DIN specification for intermodulation distortion measurement.
DIN 45405	a DIN standard describing noise measurement, virtually identical to CCIR 468.
DIN 45507	a DIN specification for wow and flutter measurement, virtually identical to IEC 386.
dither	low amplitude noise, at approximately the one-half to one LSB range in amplitude, added prior to a signal quantization in order to reduce distortion, improve linearity, and extend the available dynamic range downwards below that of an undithered system of the same number of bits.
Dolby	Dolby Laboratories, a company manufacturing and licensing noise-reduction devices and codecs for reduced bit-rate and multichannel transmission.
DSP	digital signal processor. A specialized microprocessor designed for highly efficient processing (filtering, FFT, etc.) of digitized analog waveforms. Also, digital signal processing, however implemented.
DUT	a common abbreviation in the test and measurement field for “device under test.” Also EUT.
dynamic range	the difference, usually expressed in dB, between the highest and lowest amplitude portions of a signal, or between the highest amplitude signal which a device can linearly handle and the noise level of the device.
E
EBU	European Broadcasting Union. An organization with offices in Belgium and Switzerland, which (among other things) creates standards and recommendations for broadcast-related testing.
EIA	Electronics Industries Association
EIAJ	Electronics Industries Association of Japan
elliptic filters	a family of filters possessing the sharpest possible rolloff characteristics for a given order of complexity. Elliptic filters differ from Chebyshev filters in having stopband zeros.
emphasis	signal processing before and after a noisy medium or process with the goal of improving the overall signal-to-noise ratio without affecting frequency response. Emphasis systems consist of a preemphasis filter (usually high frequency boost) before the noisy section and a complementary deemphasis filter (therefore usually cutting high frequencies).
EQ	equalization, or an equalizer.
equalizer	circuitry or equipment which produces a varying (usually adjustable) amplitude as a function of frequency.
EUT	a abbreviation for “equipment under test.” See DUT.
even order distortion	Even order distortion products are produced by nonlinearities mathematically described by even value exponents. These nonlinearities produce an asymmetrical shape in the output versus input transfer characteristic of the device. Examples include 2nd harmonic distortion and difference frequency intermodulation distortion.
expander	a signal processing amplifier which has greater gain at high input amplitudes than at low inputs, producing an output dynamic range greater than the input dynamic range. Normally used as part of an overall system with earlier compressors, to restore the original dynamic range of a signal which was reduced in a compressor.
F
fader	a variable attenuator in a mixing console or similar device, providing operator control of the level of a signal.
FFT	fast Fourier transform, a technique to compute the amplitude versus frequency and phase versus frequency information from a set of amplitude versus time samples of a signal.
Filter, band reject	see Band reject filter.
Filter, bandpass	see Bandpass filter.
Filter, high pass	see High pass filter.
Filter, low pass	see Low pass filter.
Filter, notch	see Notch filter.
Filter, weighting	see weighting filter.
flat	constant gain or attenuation across a frequency band, unfiltered.
Fletcher-Munson	researchers in the early history of audio perception, who measured and documented curves of typical frequency response of the human hearing system at various sound pressure levels.
flutter	variations in frequency of an analog-recorded and reproduced signal due to short-term variations in speed of the recording and/or playback mechanism. Flutter differs from wow in that flutter consists of speed variations at a relatively rapid rate (perhaps above 6 Hz), while wow describes variations at lower rates. Wow and flutter specifications include IEC 386, DIN 45507, CCIR 409-3, NAB, ANSI C16.5, and JIS 5551.
frequency domain	a means of representing a signal as a plot of amplitude (normally on the vertical axis) versus frequency (normally on the horizontal axis). Spectrum analyzers represent signals in the frequency domain.
Frequency shift keying	a means of transmitting digital data by shifting the frequency of a carrier signal between two values representing digital one and digital zero.
FSK	see Frequency shift keying.
fundamental	the lowest frequency component (normally also the highest amplitude) of a periodic signal.
fundamental rejection	the amount, usually expressed in dB, by which a THD+N analyzer rejects the fundamental component of the input signal. The lowest measurable distortion of a THD+N analyzer is limited by fundamental rejection, along with several other attributes.
G
gain	gain is the ratio of output signal level to input signal level.
gain set	a calibrated audio-frequency attenuator, typically with rotary-switch-controlled attenuators in 10 dB, 1 dB, and 0.1 dB steps and a level indicator.
gain, power	the ratio of output signal power to input signal power of an amplifier.
gain, transducer	the ratio of output power level than an amplifier will deliver to a load of specified resistance, to the power level that the amplifier’s driving source will deliver to a specified resistance equal to the nominal input resistance of the amplifier.
gain, voltage	the ratio of output signal voltage to input signal voltage of an amplifier.
gate	a switch or other device which controls the passage of a signal, such as is used in a tone burst generator.
Gaussian distribution	a symmetrical probability distribution characteristic of typical random noise. The distribution follows the classic bell-shaped curve where 2/3 of the values are equal to or less than the rms signal value.
go/no-go	see pass/fail.
graphic equalizer	an equalizer in which the gain at each portion of the spectrum is controlled by a separate fader. The faders are typically vertically-mounted slide controls, so that the positions of the control knobs forms an approximation to the frequency response of the equalizer.
ground loop	an inadvertent signal path formed when interconnecting the chassis of two or more pieces of equipment, each possessing a safety ground. Ground loops can cause hum-related interference.
group delay	the relative time delay between different spectral portions of a signal.
H
harmonic	a spectral component at an exact integer multiple of a fundamental frequency.
high-pass filter	a filter which passes all frequencies above a specified value essentially without attenuation, while attenuating frequencies below that value.
hot pin	the pin of a normally-balanced connector which is selected to the “high side” (ungrounded) conductor when the connector is used to carry unbalanced signals.
hum	interference at power mains-related frequencies. Hum directly at the mains frequency and odd harmonics is characteristic of magnetic coupling into the affected system. Hum at the second harmonic of the mains frequency is typically caused by inadequate filtering of power supplies with full-wave rectifiers.
I
IEC	International Electrotechnical Commission, a body responsible for preparing and publishing international standards for the electrical and electronics fields. The IEC is based in Geneva, Switzerland.
IEC flutter	the method of measuring wow and flutter described in IEC 386. Virtually equivalent to DIN 45507.
IEC60958	See digital audio signal, bi-phase coded.
IEEE	Institute of Electrical and Electronics Engineers, a professional and standards-setting organization with headquarters in New York City.
IEEE-488	an IEEE standard for interconnection of test and measurement instruments plus a computer/controller into an automated measurement system.
IHF	Institute of High Fidelity, an organization which establishes standards for the testing of consumer audio equipment.
IMD	intermodulation distortion.
ISO	International Organization for Standards, the largest of the many international groups for technical and industrial cooperation. The ISO is based in Geneva, Switzerland
ISO-MPEG standards	standards defining three “layers” of performance for low-bit-rate audio. Layers one and two are MUSICAM.
J
JIS	Japanese standard for measurement of wow and flutter using an rms detector and very long time constants in the detector filter. The JIS standard typically produces lower numbers than the NAB or IEC/DIN standards for any specific tape machine or turntable. The JIS standard is used primarily for consumer equipment.
jitter	the undesirable cycle-to-cycle variation in the period of a reference clock, such as are used in digital audio converters. Jitter can cause modulation sidebands and noise if converters operate from a jittered clock. Excessive jitter in an interface can cause digitally-interfaced equipment to malfunction.
K
L
level	the magnitude of a signal, which may be expressed in a wide variety of units such as volts, dBm, dBu, watts, etc. Also amplitude.
Limiter	a signal processing amplifier whose gain is sharply reduced above some critical threshold, so that the output signal will not exceed a specified value regardless of input amplitude.
limiting	the action of a limiter in preventing output levels above a specified value. Also, the action of a conventional amplifier when driven beyond its linear range. In this latter context, “limiting” and “clipping” are sometimes used interchangeably except that clipping may imply a more abrupt characteristic than limiting.
limits testing	testing in which measurements are compared to acceptable values so that a pass/fail decision may be made. Also called pass / fail testing.
line (FFT)	see bin.
line level	a relatively high amplitude range suitable for transmission of audio signals. Line level is typically in the 0 dBu to +8 dBu range.
longitudinal balance	an alternative method to common mode rejection ratio for measuring the degree of balance of a transmission line.
low-pass filter	a filter which passes all frequencies below a specified frequency essentially without attenuation, while attenuating frequencies above that value.
LSB	least significant bit. The bit in a binary word representing the smallest possible value change.
M
Masking, frequency	the pyschoacoustic effect where a strong signal causes weaker signals nearby in frequency to be inaudible.
Masking, temporal	the psychoacoustic effect where a strong signal causes weaker signals occurring just before or just after the strong signal to be inaudible.
matching impedances	Matching assuring that the impedances of sources and loads connected together in a system are equal, so as to assure maximum power transfer.
maximum length sequence (MLS)	a pseudorandom noise sequence designed such that every possible bit combination occurs once during each repetition cycle. An MLS sequence has the property that if it is passed through a linear device under test and a cross correlation computed between the output and input of the device, the result is the impulse response of the device. The MLS is used in quasi-anechoic acoustical testing.
maximum output level (MOL)	the maximum output level that a device can deliver without exceeding a specified distortion value. Most commonly used with analog tape recorders, where maximum output level is typically defined as maximum level achievable without exceeding third harmonic distortion of 3% for a midband signal (usually 1 kHz).
MDAC	multiplying digital-to-analog converter, often used as digitally-controlled variable resistors and attenuators.
MiniDisc	a low-bit-rate recording medium developed for consumer applications by Sony. The MiniDisc system records on a magneto-optical disk and uses the ATRAC perceptual coding system.
mixer	a specialized amplifier used in recording, broadcasting, sound reinforcement, and similar applications where a number of input signals must be combined with individual control of the level (and sometimes frequency response) of each input. Also commonly called console, mixing console, mixing desk.
MLS	see maximum length sequence.
MOL	see maximum output level.
MPEG	Motion Pictures Expert Group, a working group in the ISO setting standards for low-bit-rate video and audio.
MTS	Multichannel television sound, a generic term for the incorporation of more than one audio channel with television transmission.
multitone	testing techniques with stimulus signals consisting of more than one sine wave. Most multitone techniques use 15 or more sine waves distributed across the audio spectrum.
MUSICAM	a perceptually-based low-bit-rate encoding and decoding technique. MUSICAM complies with the ISO-MPEG standard.
MUX	multiplexer; a switch selecting among several signal paths.
N
NAB	National Association of Broadcasters, a professional-industrial organization of U.S. broadcasters which, among other things, establishes standards.
NICAM	Near-Instantaneously Companded Audio Multiplex; a technique for audio bit rate reduction developed by the BBC in the early 1980s. NICAM is not a perceptually-based system, but uses block floating point techniques.
noise gate	a signal processing device whose output is disabled (infinite attenuation) when the input signal level falls below a critical threshold, so that noise in the absence of significant program material is not passed to the output.
noise reduction	a system, normally used with recording or transmission, which reduces overall noise by processing the signal prior to recording or transmission and again following transmission or a playback. Processing before recording or transmission is normally some combination of compression and high frequency boost. Processing after transmission or at playback is the opposite of the first processing in order to restore the initial dynamic range and frequency response, while reducing the effects of noise introduced in the recording or transmission medium.
nonsymmetrical circuits, devices	unbalanced circuits or devices.
notch filter	a band reject filter with a narrow rejection band, often used to eliminate the fundamental frequency for THD+N measurements or to reject a specific spectral component such as power mains hum or a feedback frequency in a public address system.
NTC	(1) negative temperature coefficient; the characteristic of the primary electrical parameter of a component decreasing with increasing temperature. (2) a resistor possessing a negative temperature coefficient, commonly used for temperature compensation.
Nyquist frequency	1/2 the sample rate in a digital system, the frequency above which signals cannot be unambiguously coded.
O
octave	the interval between a 2:1 range of frequencies, such as 400 Hz to 800 Hz or 5 kHz to 10 kHz.
odd order distortion	distortion products produced by nonlinearities mathematically described with odd order exponents. These nonlinearities cause a symmetrical shape of the output versus input transfer characteristic of a device. An example would be third harmonic distortion, or the 13 kHz product produced by 14 kHz and 15 kHz signals (2*F1 - F2).
one-third octave	a bandwidth of 1/3 octave, or a frequency ratio of 1.2599:1. Three successive frequency changes by this ratio result in a total frequency change of 2:1 (one octave). Moderately narrow bandpass filters are often set at a bandwidth of 1/3 octave; frequency response measurement techniques that use spot frequencies (such as real-time spectrum analysis or multitone tests) often use frequencies centered at 1/3 octave distances.
oversampling	a technique used in ADCs and DACs where the sampling rate is many times higher than the minimum required for the bandwidth content of the signal. The fundamental advantage of oversampling is a simplification in requirements of anti-alias and reconstruction filters.
P
PAC	a low-bit-rate perceptual coding technique developed by AT&T.
pad	see attenuator.
Pan pot	panoramic potentiometer. A variable attenuator used to control the proportions of an input signal which appear in each of the two stereo outputs. A pan pot can thus position the stereo image of a signal from left to right.
parametric equalizer	an equalizer in which the parameters of center frequency, bandwidth (Q factor), and amount of gain or attenuation of a number of filters are all adjustable.
parasitic oscillations	unwanted very high frequency oscillations that can occur in audio amplifiers, resulting in degraded noise and distortion performance.
PASC	Precision Adaptive Subband Coding, the perceptually-based low-bit-rate coding technique used in the Philips DCC system. It is similar to the MUSICAM system standardized by ISO.
pass/fail	testing in which measurements are compared to acceptable values so that a pass/fail decision may be made. Also called limits testing.
passband	the frequency band of a filter in which signals are essentially unattenuated; also the flat response range of a digital converter between the device corner frequencies.
passband ripple	amplitude variations across the passband of a device. This term usually refers to the small-amplitude ripple seen in ADC and DAC response curves.
patch panel	an arrangement of connectors plus flexible cables and plugs, permitting a variety of audio devices to be connected to one another as desired. Commonly used in broadcasting and recording studios.
PCAQM	Personal Computer Audio Quality Measurements.
PCM	see Pulse code modulation
peak	the maximum instantaneous excursion of a signal.
peak level	Peak voltage (Vp) is the maximum instantaneous excursion of a signal. Peak-to-peak voltage (Vpp) is the maximum amplitude difference between positive-going and negative-going peaks of a signal.Peak voltage level is another common level measurement for audio signals. Peak values are typically higher readings than rms values for the same signal.
peak to peak	the maximum amplitude difference between positive-going and negative-going peaks of a signal.
perceptual coding	low-bit-rate coding of a digital signal according to an understanding of human perceptions of sound, so that the most important portion of the signal are coded with the greatest accuracy and coding capability is not wasted on portions of the signal which are inaudible.
phantom power	a system for supplying dc power from a preamplifier or mixer microphone input to a microphone, on the same cable conductors which carry the microphone’s output signal.
pilot tone	a signal (19 kHz in the case of stereo FM broadcasting) transmitted as a phase reference for use in the receiver to demodulate the double sideband suppressed carrier stereo subcarrier.
pink noise	noise whose spectral power distribution is such that the power per octave, per decade, or in any other equal-percentage section is the same anywhere across the spectrum. For example, pink noise has the same power in the octave between 50 Hz and 100 Hz as in the octave between 10 kHz and 20 kHz.
pole	frequencies (in the context of complex variable theory) at which the transfer function of a device exhibits a maximum value.
pot	potentiometer. Variable attenuator to control the gain of an amplifier.
power gain	the ratio of output signal power to input signal power of an amplifier.
PPM	(1) in test and measurement, parts per million.(2) in professional audio and, broadcasting, peak program level meter.
preamplifier	amplifier stage or stages used to bring low-level signals, such as those from microphones and phonograph pickup cartridges, up to a higher, standard level suitable for signal routing, mixing, monitoring, etc.
preemphasis	signal processing before recording or transmission through a noisy medium. Combines with later deemphasis (see) to produce overall improved signal-to-noise ratio with flat overall frequency response. See deemphasis.
processor	an amplifier designed to modify characteristics of a signal such as frequency response or dynamic range. Processors in common use in broadcasting and professional audio include equalizers, compressors, limiters, reverberation units, modulation processors, noise reduction units, noise gates, etc.
professional	In audio, often used to designate digital audio format. See digital audio signal, bi-phase coded.
proof	common abbreviation for proof of performance.
proof of performance	a term in common use among U.S. broadcasters to describe the complete end-to-end performance tests (studio microphone connector through transmitter, antenna, and demodulator) to demonstrate proper operation of a broadcast facility.
pseudorandom noise	noise whose amplitude-vs-time distribution appears to be random when examined over a short period of time, but which in fact exactly repeats a pattern of a certain duration. The spectrum of a pseudorandom noise signal has power only at the frequencies corresponding to integer multiples (harmonics) of a fundamental frequency whose period is equal to the repetition cycle duration. For example, a pseudorandom noise sequence with a two second repetition cycle will have a spectrum consisting of power at every harmonic of 0.5 Hz.
psophometric filter	a filter whose response is based on the frequency response of the human hearing system. Most noise weighting filters are psophometric filters.
psychoacoustics	the area of science combining acoustical stimulus and human response to that stimulus.
PTC	(1) positive temperature coefficient; the characteristic of the primary electrical parameter of a component increasing with increasing temperature. (2) a resistor possessing a positive temperature coefficient, commonly used for overload protection in loudspeakers and electronic equipment.
pulse code modulation	PCM. A form of data transmission in which amplitude samples of an analog signal are represented by digital numbers.
Pulse code modulation	a form of data transmission in which amplitude samples of an analog signal are represented by digital numbers.
pulse code modulation	a form of data transmission in which amplitude samples of an analog signal are represented by digital numbers. Abbreviated PCM. Almost all digital audio schemes use PCM.
pulse width modulation	PWM. A form of data transmission in which amplitude samples of an analog signal are represented by the duty factor of a pulse train. Often used in switching (Class D) amplifiers.
PWM	see Pulse width modulation.
Q
Q factor	a measurement of the selectivity or sharpness of a bandpass or bandreject filter. For a bandpass filter, the Q is equal to the ratio of the center frequency to the bandwidth at the -3dB points. For example, a bandpass filter with a Q of 5, tuned to a center frequency of 1,000 Hz, will have a 3 dB bandwidth of 200 Hz.
quasi-anechoic	an electronic technique for eliminating the effect of acoustical reflections while testing in a normal space with reflections. Quasi-anechoic techniques all involve measuring signals whose arrival time corresponds to the acoustical propagation delay of the desired direct path, while attenuating or eliminating signals which arrive later due to the longer path length of the reflection. Quasi-anechoic techniques include impulse stimulus with gating, Maximum Length Sequence, and Time Delay Spectrometry.
quasi-peak	a fast-attack, slow-decay detector circuit which approximately responds to signal peaks. Specifically, the detector response called out in CCIR Recommendation 468.
R
random noise	noise whose amplitude-vs.-time distribution is mathematically random and unpredictable, never repeating. The spectrum of a random noise signal is continuous with power at all frequencies, rather than power only at certain points as with pseudorandom noise.
RC	resistor-capacitor. Commonly used to describe filters and oscillators, as opposed to LC (inductor-capacitor).
RDAT	rotary head digital audio tape, a standard originally developed as a consumer medium for recording of digital audio signals that found wide use in professional applications. Now largely replaced by hard-disk and solid state recording.
real-time analyzer (RTA)	a simple audio spectrum analyzer consisting of a number of bandpass filters tuned to staggered frequencies across the audio spectrum, each filter followed by a separate detector and indicator. Typical real-time analyzers are made up of one octave bandwidth filters (thus about 10 bands across 20 Hz–20 kHz), 1/2 octave filters (20 bands), or 1/3 octave filters (about 30-31 bands across the audio spectrum). Real-time analyzers are commonly used with microphones and a pink noise source or program material for acoustic measurements or monitoring. They lack the selectivity, resolution, and accuracy for most electronic audio device measurements.
reconstruction filter	a low-pass filter following a D/A converter, used to remove the high frequency clock signal and smoothly integrate between the discrete voltage values from the D/A.
record (digital sampling)	a section of sampled signal in memory, represented as binary numbers.
recovery time	the time interval required following a sudden decrease in input level to a device for the output of the device to reach some stated percentage of the eventual signal level.
rectangular probability density function dither	RPDF or rectangular dither. Dither which has equal probability of occurrence at any amplitude value between plus one-half LSB and minus one-half LSB deviation from the nominal value. Thus, a graph of probability versus digital value is a rectangle.
residual distortion	the irreducible minimum distortion of an audio generator and analyzer. Residual distortion is thus the “floor” below which the instrument is not useful; measurements above but approaching within 6 to 10 dB of the residual distortion value are less accurate.
residual noise	the irreducible noise in a measurement instrument which sets a floor for amplitude measurements.
resolution	the smallest change in a measured parameter to which a measurement instrument can respond.
resolution	the smallest change in a measured parameter to which a measurement instrument can respond. The resolution of a PCM digital audio system corresponds to its digital word length; digital systems, for example, can be described as having “16 bits of resolution.”
return loss	a measure of power reflected back to the originating end of a channel due to impedance mismatches in the channel.
reverb	reverberation.
reverberation	acoustically, the effect caused by acoustical waves traveling various paths (direct path from sound source to listener location plus some number of reflections) with the reflected power being delayed in time due to greater path length. Many signal processors simulate reverberation electronically.
reverse termination	a resistance or impedance connected at the input to a device under test in order to make repeatable noise measurements.
ripple	(1) undesired ac variations on a dc power supply output. (2) variations in frequency response in the passband of a filter.
ripple	(1) undesired ac variations on a dc power supply output. (2) variations in frequency response in the passband of a filter.
rms	see root mean square.
rms level	Rms (root mean square), is the preferred form of ac signal detection which measures amplitude in terms of its equivalent power content, regardless of signal waveshape.The rms voltage level (Vrms) is the most common level measurement for audio signals, although this level is often stated in a variety of relative units such as dBV or dBu. Rms values are typically lower than peak values for the same signal.
rolloff frequency	the frequency considered the transition between the attenuated and non-attenuated portions of the frequency response of a high-pass or low-pass filter. The filter has a specified attenuation, usually either 3.01 dB or the magnitude of the ripple, at the rolloff frequency.
root mean square	the preferred form of ac signal detection that measures amplitude in terms of its equivalent power content, regardless of signal waveshape. Abbreviated rms.
RPDF	See rectangular probability density function dither.
rss	root sum square; a method for combining the power of a number of signals by squaring each, summing (adding) these squared values, and finally extracting the square root of the sum.
RTA	see Real time analyzer.
rub and buzz	a variety of distortions and noises created in a loudspeaker, mostly due to mechanical defects such as voice coil rubbing the magnet, cone touching connection wires, etc.
S
S/N ratio	see signal to noise ratio.
signal-to-noise ratio	the difference in level between a reference output signal (typically at the normal or maximum operating level of the device) and the device output with no signal applied. Signal-to-noise ratio is normally stated in dB. The device input conditions for the noise measurement must be specified, such as “input short-circuited” or with a specific value of resistance connected at the device input instead of a signal.
S/PDIF	Sony / Philips Digital Interface; a digital interface for consumer audio equipment. Sometimes also referred to as the EIAJ interface. The S/PDIF is similar to the professional AES3 interface, but is normally an unbalanced coaxial signal of lower amplitude. Most of the status byte definitions are different between S/PDIF and AES3. See digital audio signal, bi-phase coded.
sample clock	the time reference signal in a digital system which determines the intervals at which the signal will be sampled.
sample rate, sample frequency	the frequency at which the signal is sampled in a digital system. The sample rate must exceed twice the highest analog frequency to be converted. Commonly-used sample rates are 48 kHz, 44.1 kHz, and 32 kHz.
SAP	secondary audio program; a third channel in the BTSC stereo system which may be used for multilingual broadcasting or other services.
saturation	the effect occurring when driving an audio device (particularly an analog magnetic tape recorder or transformer) beyond its linear region, where the output amplitude cannot linearly follow the input. Used somewhat interchangeably with the terms “limiting” and “clipping.”
SCA	an auxiliary multiplexed channel or channels in a broadcast FM signal, located higher in frequency than the stereo sub-channel.
SDIF	Sony Digital Interface, a serial digital interface using three coaxial cables. The SDIF and SDIF2 (a later version) were originally used on the Sony PCM-1600 series of digital audio processors.
second order products	distortion products produced by a term with the exponent “2” (i.e., a squaring function) in the device transfer function.
SEDAT	a low-bit-rate coding technique developed by Scientific Atlanta.
separation	the isolation (usually stated in dB) between channels of a stereo or multichannel device. Crosstalk.
shaped dither	dither in which the power-versus-frequency distribution has been modified to place most of the power at frequencies where the human hearing system is not as sensitive.
shielded cable	a construction technique for cables in which a metallic outer conductor completely surrounds one or more signal-carrying conductors. This outer shield is normally connected to ground (earth) at one or both ends.
short circuit	zero-Ohm connection.
sigma-delta	an ADC technique characterized by a relatively low-resolution quantizer operated at very high speeds, digitally low-pass filtered to obtain the final digital signal.
Signal-to-noise ratio	the difference in level between a reference output signal (typically at the normal or maximum operating level of the device) and the device output with no signal applied. Signal-to-noise ratio is normally stated in dB. The device input conditions for the noise measurement must be specified, such as “input short circuited” or with a specific value of resistance connected at the device input instead of a signal.
single-ended	unbalanced.
skirt (filter)	the portion of response curve of a bandpass, high-pass, or low-pass filter where the attenuation increases at a rapid rate.
slew rate	the rate of change of a signal amplitude, typically expressed in Volts per microsecond.
slew rate limiting	the condition when an amplifier output is unable to follow the rate of change of the input signal.
slider	a particular design of attenuator or potentiometer in which the control know travels in a straight line (as opposed to rotary attenuators).
SMPTE	Society of Motion Picture and Television Engineers. A professional and standards-setting organization.
SMPTE RP120	a specification for intermodulation distortion measurements.
SMPTE 276M	See digital audio signal, bi-phase coded.
spectrum analyzer	a measurement instrument which measures and displays a signal as a frequency domain representation of amplitude (usually vertical scale) versus frequency (horizontal scale). Spectrum analyzers may be designed with a variety of technologies including real-time analyzers (bandpass filter bank), heterodyne or scanning analyzers with a local oscillator, mixer, and selectivity at a fixed intermediate frequency, scanning tunable filters, and FFT analyzers.
split site	a broadcasting operation with studio at one location and transmitter at another. Testing of split-site systems imposes some constraints on test equipment and techniques because signal must be injected at one location while measurements are made at another.
status bits	metadata carried in professional and consumer digital audio bitstreams. See digital audio signal, bi-phase coded.
stereo generator	a device which converts left and right channel audio into a composite multiplexed signal suitable for modulating an rf carrier.
stimulus	a known signal applied to an audio device under test, chosen to stimulate device behavior useful for the test being performed.
STL	see studio to transmitter link.
stopband	the frequency band across which a filter has at least some minimum specified value of attenuation.
studio to transmitter link	a transmission path from program material from broadcast studio to transmitter, usually a microwave link.
sum product	an intermodulation distortion product at a frequency equal to the sum of two input frequencies.
sweep	A sweep shows one parameter as a function of a second parameter. The second parameter is moved (swept) across a range of values. Results are displayed as X-Y graphs or as a table of values. In audio measurement, many sweeps display level or distortion as a function of swept frequency. Amplifier testing often requires level sweeps, displaying output level or distortion as a function of swept input level.
symmetrical	balanced
T
TDS	see time delay spectrometry.
terminating	connecting the specified load resistance or impedance to a device.
terminating	connecting the specified load resistance or impedance to a device.
termination	a specific resistance or impedance value which must be connected to the output or input of a device under test for certain parameters to be measured.
THD	THD stands for Total Harmonic Distortion, a measure of all the harmonic distortion products in the DUT’s output, with the fundamental stimulus tone removed, and without consideration of noise. THD measurements are usually made with band-limiting or weighting filters. True THD measurements can be valuable but are rare, since with most techniques it is difficult to measure the distortion products without also measuring the noise. See THD+N.
THD+N	THD+N stands for Total Harmonic Distortion plus Noise, a measure of everything in the DUT’s output, with the fundamental stimulus tone removed. Noise and other interfering signal such as hum, buzz and aliased high-frequencies are reported in the result. THD+N measurements are usually made with band-limiting or weighting filters. THD+N measurements are regarded as a key benchmark of a system’s performance; the measurement is easy to perform and widely understood, and it shows at a glance not only distortion performance but also indicates a system’s immunity (or lack of it) to noise and other interfering signals.
third octave	see one-third octave.
third order	distortion products produced by a cube (exponent of 3) term in a devices's nonlinear transfer function.
three head	a tape recorder design including erase, recording, and playback heads (in that order) so that a just-recorded signal may be monitored from the tape during the recording process.
TIM	transient intermodulation distortion. See DIM (dynamic intermodulation distortion).
Time delay spectrometry (TDS)	A technique for quasi-anechoic frequency response measurements, in which a sine wave stimulus is linearly swept in frequency while a bandpass filter in the analyzer is swept at the same rate, but delayed in time by the acoustical propagation delay between the loudspeaker under test and the measurement microphone. As a result, the direct (first arrival) signal is passed unattenuated while reflected signals are attenuated since they arrive later (due to longer path length) when the filter has swept onwards to another frequency.
time domain	a means of representing a signal as a graph of amplitude (usually on the vertical axis) versus time (on the horizontal axis). An oscilloscope produces a time domain representation of a signal.
toneburst	a succession of sinewaves preceded and followed either by no signal or by the same sine wave frequency at a lower amplitude. Tonebursts are used to simulate program material during calibration of peak meters and in performance tests of audio processors, and for transient response testing of loudspeakers.
Toslink	See digital audio signal, bi-phase coded.
TPDF	See triangular probability density function dither.
transducer	a device to change acoustical power into an electrical signal (microphone, for example) or electrical power into an acoustical signal (loudspeaker, for example).
transducer gain	see gain, transducer.
triangular probability density function dither	TPDF dither or triangular dither. Dither in which a graph of the probability of occurrence of an amplitude rises linearly from zero at values plus-or-minus one least significant bit above or below the nominal value, to unity at the nominal value. The graph thus forms a triangle of unity value at the horizontal center (nominal digital value on the horizontal digital value scale), falling to zero at one LSB to the left or right on the horizontal scale.
trigger	an event which causes another event or action, often initiating a signal generation or acquisition.
tweeter	a loudspeaker unit optimized for performance at high (treble) frequencies.
twisted pair	an audio cable consisting of two similar conductors twisted together to help assure that any noise voltages coupled into the cable will be in equal amplitude between each conductor and ground (earth), therefore representing a common mode signal which will be substantially rejected by a high quality balanced input.
two sigma flutter	the value of wow and flutter which is exceeded by only five percent of the instantaneous readings over a period of time.
two track	tape recorder for stereo recording and/or playback.
U
unbalanced	an audio connection in which the desired signal is present as a voltage with respect to ground or common, rather than as a differential signal across a pair of balanced conductors.
V
VCA	see voltage controlled amplifier.
VCO	see voltage controlled oscillator.
voltage controlled amplifier	an audio amplifier with a dc voltage control port which electronically varies the gain of the amplifier.
voltage controlled oscillator	an oscillator whose frequency can be varied via a dc control voltage.
voltage gain	see gain, voltage.
VU meter	a volume unit meter, used to indicate program levels in broadcasting, recording, and similar applications.
W
weighting filter	a filter with varying attenuation as a function of frequency so as to produce a measurement where the various spectral components affect the measurement in a specified fashion. Most commonly-used weighting filters are attempts to correspond to the varying response of the human hearing system in order to produce measurements (usually of noise) which correlate well with human observations.
white noise	noise whose spectral power distribution is such that there is equal power per Hz, anywhere in the spectrum. For example, white noise will have the same power in the 30 Hz bandwidth between 70 Hz and 100 Hz as in the 30 Hz bandwidth between 10,000 Hz and 10,030 Hz.
WHQL	Windows Hardware Quality Labs. Microsoft’s program to ensure hardware compatibility by recommending standards and implementing compatibility testing and verification. On the Web at www.microsoft.com/hwdq/hwtest/default.asp
WHQL PC2001	The common name referring to the Intel/Microsoft document PC 2001 System Design Guide.
window	an amplitude-vs.-time function used to multiply the corresponding samples of a digitized waveform before computing an FFT. Window functions go to zero at the two ends of the record. This provides better selectivity, reducing the signal spreading when non-windowed FFTs are computed on signals which are not exactly synchronous in the signal buffer length being transformed.
woofer	a loudspeaker unit optimized for performance at low (bass) frequencies.
word length	the resolution of an audio coding system, expressed as the length of the digital word in bits. Each bit adds approximately 6 dB to the theoretical dynamic range available. Thus, a 16-bit digital system is capable of approximately 96 dB dynamic range. In high-performance linear PCM, word length is typically 16, 20 or 24 bits. Telecom and compressed coded audio may use fewer bits, with word lengths of 14, 13, 12 or 8 bits. Also called “word width” or "bit depth."
wow	a low frequency variation in pitch of an analog recorded and reproduced signal due to low rate speed variations of the mechanical system (disk or tape). See also flutter. Wow and flutter specifications include IEC 386, DIN 45507, CCIR 409-3, NAB, ANSI C16.5, and JIS 5551.
X
XLR connector	a high-quality balanced connector designed for audio applications. Sometimes also called a Cannon connector.
XML	Extensible Markup Language (XML) is a simple, very flexible text format derived from SGML (ISO 8879). Originally designed to meet the challenges of large-scale electronic publishing, XML is also playing an increasingly important role in the exchange of a wide variety of data on the Web and elsewhere.
Y
Z
Zero	frequency (in the context of complex variable theory) at which the transfer function of a device has zero value.

Technotes & Appnotes

Technotes and Appnotes look at various topics in audio test and measurement—sometimes a particular solution, perhaps a time-saving tip or an in-depth exploration of a difficult concept. Technotes are shorter documents that tend to look at a particular issue or problem, while Appnotes tend to examine larger sets of audio test issues, for example, all the aspects of measuring a CD player performance.

Many of the notes are specific to Audio Precision analyzers, and some to a particular instrument. However, the approaches and strategies taken in a certain testing environment are often transferable to another situation, and our customers find value even with the older notes written for legacy instruments.

• Read abstracts for each Technote and Appnote

 

For general audio research

Whitepapers
Audio Precision co-founder Bruce Hofer and several other AP engineers have contributed many white papers to the audio engineering community. They are presented here for your reference.

• Read abstracts for each whitepaper

Audio training resources
AP has several online tutorials courses as well as DVDs and books available to help users new to audio understand basic concepts through to qualified engineers improve their skills

• View available training resources

Audio.TST back issues
Audio.TST is a monthly publication that looks at the current issues in audio test and measurement. Subscription is free.

• View the back catalog of Audio.TST
• Subscribe to Audio.TST