CD Audio Recording Glossary

A device used top convert analogue signals into digital (discrete time) signals. These digital signals can then be processed by computers. Computer sound cards usually contain an A-D converter and a D-A converter (Digital to Analogue) The better quality cards now have 24 bit processing and 16 Bit is the standard for CD Audio.

Automatic Gain control or Automatic Level Control is designed to hold system gain at a relatively constant value independent of the incoming audio, Audio Signals below a certain level are upwardly expanded and signals above the threshold are compressed.

An electronic method for increasing the level of signal in a circuit.

In audio terms is usually expressed as a relative term (ratio of 2 levels) in dB – Decibels. Although you may hear of it in reference terms as Volts or Sound Pressure Level (SPL).

Reduction of an audio signal. Forensic filters are designed to reduce the amount of unwanted audio signals by large amounts, while retaining critical voice information

Generally accepted as 20 Hz to 20,000 Hz (20KHz) for humans. A high quality audio replay or recording system should be able to reproduce this frequency range within +_ 3dB.

Analogue tape recorders (such as cassette or reel to reel ) rely on recording and replay heads to transfer information to and from the tape. These heads should be in alignment and perpendicular to the tape. (this is called the head azimuth). If the heads are not correctly aligned replay of the audio will be degraded, typically resulting in a loss of high frequency and can also result in phasing of the signal in stereo or multi-track applications.

A filter that allows a selectable range of frequencies to pass without attenuation. Band pass filters can have variable widths and filter shapes. The filter shape will affect audio outside of the filter range in different ways.

Typically in Digital audio a buffer is a section of memory that is used as temporary storage during input and output of the signal. “Buffer under runs” can sometimes be seen in transferring audio to CD. This generally results in a poor or failed recording.

Audio signals that are usually harmonically related to the mains frequency. Buzz generally has a larger amounts of high frequency harmonics compared to “hum”.

An 8-bit word. Each sample of a mono wav file is represented by two 8-bit bytes. 2 x 8-bit bytes are used to represent audio in a 16-bit recording.

(1 kilobyte =1,024 bytes).

In music 1,200 cents = 12 semitones = 1 octave ( “Just intonation” scale).

Optical Digital storage devices; typically 120mm in diameter and 1.2 mm thick.

Made from high-grade polycarbonate compound. Used to store audio, video, text & photographs. The data is read via an infra red laser diode of approx 780 nM.

Audio clipping occurs when the signal value exceeds the headroom in a recording or playback system. The result of this clipping is heard as distortion and can vary from “pleasant” to unbearable. Clipping can be visually observed as a flattening of the tops of the audio signal in a wave form display.

An audio signal device that reduces the dynamic range of the incoming audio.

Often used to increase the “average” level or loudness of the audio without clipping the peaks.

Short term impulse noises found on 78, Edison cylinder or vinyl phonograph recordings. Can also occur in the digital realm with tape drop out, poor tape replay or signal interference.

In stereo or multi-track audio this is the amount of unwanted signal that bleeds from one channel to another.

dB (decibel)

1/10 of a Bel . A Bel is the basic unit for the measurement of sound intensity.

Converts digital signals back to the analogue domain so that they can be played back on compatible equipment.

Reverse process of “Pre-Emphasis”

Reduction of distortion of the audio signal caused by excessive sibilance.

Generally unwanted changes to the audio signal. Various types of distortion can be found including, Harmonic, Inter-modulation, Clipping, Crossover, Phase & Jitter and amplitude distortion.

With an audio signal it’s theoretically the ratio of the smallest to its largest level.

The attack, release, sustain and decay of an audio signal defines it’s envelope and average amplitude.

Performs the opposite function of a compressor. Increasing the dynamic range of an audio signal above a predefined threshold. “Companding” is the usage of both compression and expansion in the transmission of a signal, often done to improve signal to noise ratio.

Rapid frequency modulation of an audio recording or signal due to the changing velocity of the record or tape. Occurs in the range of 6-250Hz.

A signal processor that divides the audio signal in to a number of bands so that the tonality can be adjusted. Each band can be increased or decreased in volume. Typical graphic equalisers are 10 band octave based, 20 or 30 band with 1/3 octave control range.

A ground loop occurs when there is more that one potential path for the signal to return to earth. Quite often a ground loop will inject a hum in to the audio signal.

Odd and even multiples of the fundamental signal frequency. The characteristic of an instrument or voice is created by the unique distribution of harmonics.

The Hertz (Hz) is the fundamental unit of frequency. 1Hz = 1 cycle per second,

1 Kilo Hertz = 1,000 Hz or 1 KHz.

The combination of 2 dissimilar frequencies causes several other frequencies to be produced based on the sum and difference of the two original signals. Sometimes also called “beat frequencies”

Random high frequency noise at the top end of the audio spectrum. Typically above 5KHz. Often found on audiotape or cassettes or poorly recorded digital projects.

Hum

This is generally low frequency noise introduced into a recording or sound

system, that is harmonically related to the power frequency. In Australia and Europe, this will be 50Hz in the USA 60 Hz. in both cases it will include harmonics of the line frequency. So you will hear 100Hz, 120Hz, 150Hz 180Hz etc etc. "Hum" frequencies are the fundamental (usually due to ground loops) and /or its second harmonic (due to defective power supply filter capacitors in electronic equipment).

Impedance (Z)

In an A.C circuit this is the total opposition to the flow of current Impedance is measured in Ohms

Wherein: Z = Impedance in Ohms.

Careful matching of input and output impedances in audio equipment will ensure optimum transfers, recording and playback of audio signals.

IPS (Inches per Second)

The linear velocity of magnetic tape moving past a recording or playback head is referred to in terms of IPS (inches per second).

KHz (Kilo Hertz)

The unit used in the measurement of frequency equal to 1000 Hertz. In earlier

times, this term was Kilocycles (1,000 cycles per second).

Least Significant Bit (LSB)

The smallest quantised increment which an Analog to Digital or Digital to Analog converter can resolve an analog Voltage or current.

Limiter

Similar to a compressor a Limiter but with almost no allowance for signals to pass through above a set threshold. Typically settings of a limiter will be 20:1 or above. That is it takes at least a 20db increase in the input signal to create a 1db increase in the output level.

Loudness

Unit of measurement is the SONE . It represents the loudness of a sound that is the perceived magnitude due to the auditory sensation

produced by an acoustic signal. A function of frequency and signal amplitude whereby different frequencies need different amplitudes to be perceived as having the same loudness.

A filter that attenuates all frequencies that fall above its corner frequency (or centre frequency). The degree of attenuation of a signal outside of the

filters range depends on the frequencies concerned , the corner frequency, and slope (order) of the low-pass filter. This type of filter is often used to reduce the hiss on a recording. However, low-pass filters will also attenuate the "high frequencies" on a recording at the same time, which make them generally undesirable for this application.

MByte

One million Bytes. (Sometimes 1024 KBytes for disks)

Milliseconds

The unit used in the measurement of time equal to 1/1000th of a second.

Monophonic

An audio signal or a Wave file that contains only one unique channel of aural

information is sometimes referred to as being monophonic or mono.

There are two relatively common musical scales. They are the “Scale of Just

Intonation”, and the “Scale of Equal Temperament”. The Scale of Just Intonation requires at least 30 discrete frequencies for each octave, making it relatively impractical to build musical instruments with fixed tones to play in the Just Scale. Therefore, the scale of Equal Temperament containing only 12 notes per octave is the one in general use.

Noise

Unwanted disturbances superimposed upon a useful audio signal that tends to obscure its information content. See “Signal to-Noise ratio” for more information.

A noise gate is an electronic circuit, which turns off an audio signal when its input signal is below a predetermined threshold value.

Notch Filter

A filter, which attenuates all frequencies close to the center frequency of the filter setting. The degree of attenuation and the range of frequencies which are attenuated by this filter are determined by the filters Q or bandwidth. This type of filter is often used to minimise hum or acoustic feedback from a recording.

Octave

An octave is a group of eight musical notes and also a doubling of frequency. For example, the range of frequencies from (A) 440 Hz to 880 Hz is 1 octave. The next octave will end at 1760 Hz. Note that in two octaves, the frequency has increased by a factor of four.

A variable electronic “tone” filter in which the following three parameters may be adjusted on each parametric channel:

1. Frequency

2. Level (attenuation or amplification)

3. Bandwidth

Parametric equalizers can be mono or 2 channel and usually contain several filters for each.

Pink Noise is random electronic noise, which is characterised as containing equal energy per unit octave.

Pre-Amplifier

(Pre-amp)

A device that provides Voltage amplification of an audio signal. Sometimes contains various inputs for a variety of signal sources and tone controls

Pre-Emphasis

The intentional added amplification which is sometimes applied to the top end of the audio spectrum during a recording or radio transmission process in order to raise the signal level at high frequencies substantially above the noise level of the system. This process is reversed during the reproduction

process of the signal in order to recreate an overall flat frequency response. The result of this process is an improvement in the signal-to-noise ratio of the system.

The minimum amplitude increment into which the A-D converter of a discrete time system can divide an analog signal. Many systems are 16 bits, which is 1 part in 65,536. However, many systems now support up to 24-bit I/O resolution. Resolution is also referred to as the minimum "time slice" into which a sampled signal is divided or displayed. For example 44.1Khz (44,100Hz) is the resolution of standard stereo CD Audio.

Reverberation

Electronic Reverberation is The process whereby the acoustical reflections of a room or concert hall are reproduced artificially, with devices such as

tapped delay lines working in conjunction with mixing and phase shifting devices or algorithms. Many stand alone Reverb devices exist and software reverb “plug –ins” are now becoming very common with computer based recording systems.

RIAA “Equalisation Curve”

Record Industry Association of America

The RIAA Curve is an equalisation frequency response contour which was

adopted by some manufacturers of LP records around 1955. And which became a standard practice around 1960 RIAA/IEC Curve is a 4 band variation of this curve.

RMS (Root Mean Squared)

Often described, as the average volume of a signal or it’s “perceived loudness”. RMS is a percentage of the instantaneous peak value of a signal.

Roll-off

Roll off is the attenuation or reduction of an audio signal. In the record industry, roll-off usually refers to the amount of attenuation in dB @ 10

KHz which must be applied during record playback in order to achieve a flat response on the high end of the audio spectrum. For example, the roll-off for the RIAA curve is –13.7 dB and –12 dB for the AES curve.

RPM (Revolutions Per Minute )

Some common record speeds are 33.33 RPM for LP's, 45 RPM for records with the same name. A variety of speeds are encountered with other formats such as 78’s and Wax Cylinders.

Rumble is a low frequency noise signal, typically below 50 Hz, which is often found on records.

The rate at which an analog signal is converted to discrete numbers by an A-D converter. For audio systems, sample rate is expressed in KHz. Common sample rates include:

1. 11.025KHz (Telephone systems)

2. 22.05 KHz (CDROM Audio)

3. 44.1 KHz (Std. Stereo CD Audio)

4. 48.00 KHz (Pro Audio/Video Recorders)

5. 96.00 KHz (DVD Audio & Multi-track recorders)

6. 192.00 KHz ( High End Audio Recording systems, SACD and DVD-A).

Signal-to-Noise Ratio

The ratio of signal-to-noise (Voltage, current, or acoustical sound pressure level) that is expressed in dB.

Single-Ended Noise Reduction is a process where noise is removed from an un-encoded audio signal. Filters such as Impulse, Continuous, Dynamic,

Harmonic Reject, and Notch filters are examples of Single-Ended Noise Reduction tools.

Sone

A Sone is a unit of measurement for sound loudness.

A Spectrograph is a system for presenting audio data in a graphical form and is a special case of a spectrum analyser coupled to an oscilloscope.

THD (Total Harmonic Distortion)

THD is a value as to how much non-linearity a system is imposing upon an audio signal or signal path.

Transformer

An alternating current device used to impedance match transducers and electronic circuits to one another. Sometimes, these devices are used with a unity turns ratio to provide isolation from one circuit to another rather than to impedance match the two. This is useful in audio applications when it

is necessary to break a ground loop source of hum or noise in a system.

Wave file

A common audio file format used in Windows operated systems. This (.wav) file format can be used at a number of different sampling rates. Stereo 44.1Khz is equivalent to Standard CD Audio.

White Noise

White Noise is random noise that is characterised as containing equal energy per unit frequency (Hertz).

Wow

A slow periodic change in the pitch or low frequency flutter which may be present on phonograph, tape, or soundtrack recordings due to a non uniform velocity of the recording medium. Wow is generally a frequency modulating effect that occurs at a deviation rate between 0.5 to 6 Hz.

Wow and Flutter

Wow and flutter is the combined FM effect of both mentioned parameters. The frequency spectrum in which this rate of frequency deviation is made is in the spectrum that exists between 0.5 to 250 Hz.

This CD Audio Terminology & Glossary has been compiled over many years from various sources. If you would like to add a subject to the Audio Glossary, or suggest alterations to the Glossary's terminology, please contact us.