Wed 19/4/2000 Audio Mastering - A Neglected Art Music & Sound for Multimedia & INTERNET.

Wed 19/4/2000 www.MusicSound.net 1

Audio Mastering - A Neglected Art

Music & Sound

for

Multimedia & INTERNET



A much neglected process of audio engineering in the music business. It is a process in which the engineer requires in-depth technical knowledge, musical and creative skills, and sensitive and well-trained ears. Audio mastering is both an art and a science: the art of music with the science of sound.



Why Mastering is necessary ? (eg. Demo & Master)

Analog & Digital Broadcasting, DVD & CD-Audio

What is Mastering ?

Audio mastering is the linking stage between the recording and mixing stages of music production, and product manufacture (CD duplication etc). The aim is to take a selection of mixes from the recording studio, which may have been recorded months apart, and with different equipment, and transform them into a smoothly flowing album of professional quality, which can then be commercially released.



According to Richard Wentk at Future Music Magazine

The golden rule of mastering is to remember that -

“Mastering doesn’t make anything sound different, it just makes it sound “better”….. You know it when you hear it on professional music – and that’s the result you want to create for yourself ; no big changes, just a better sounding bass, more detail in the mid, and a natural treble sparkle. In fact the best mastering sounds completely natural, like the music always wanted to

sound, and all you’ve done is help it along.”



The process of mastering happens within three steps:

Final mixes on a format such as Digital Audio Tape (DAT)

Arrange the songs into the desired sequence, with adequate spacing between each song.

Next the engineer uses his unique mastering skills to coax the best possible sound out of each song.



Specialist processing techniques:

Error correction

Spectral balance

Unification of the sonic characteristics of each song

Creating a consistent listening level

Transferred onto a format to be sent to the manufacturing facility. This “production master” will be the finished album exactly as it will be heard by the consumer.



Why do we need a separate Mastering facility ?

Special equipment and listening environment is needed.

Why do we need a separate Mastering Engineer?

A successful mastering engineer must have a musical as well as technical background. They must have a good sense of pitch and

timing, as well as extremely reliable equipment.



Analog or Digital -when to convert?

Audio CD mastering at many high-end mastering houses is still performed with analogue equipment in the chain. They use very expensive and often custom-made equipment. Some engineers say that there are some effects that are only possible in the analogue domain.

Digital CD mastering also has it’s advantages and effects that cannot be found in the analogue domain, for example, the cut, copy and paste functions in hard disk editing, or some forms of audio restoration. New ways are constantly being found to simulate the well loved sound of analogue audio, in the digital domain.



Special techniques for digital processing:

“Microdynamic enhancement” - is a technique, which is claimed to restore the liveliness and life of a great live recording or simulate it in a studio recording. It can be applied to overly compressed recordings or midi-sample dominated recordings to give them life. It claims to help remove the veils introduced in multi-generation mixdowns, tape saturation and sound “shrinkage” which comes from using too many op-amps in the signal path.



Special techniques for digital processing:

“Stereoisation” - is a technique which is claimed to alter the width on flat-sounding material. It appears to capture and bring out the original ambience in a source. It is used when recovering what may have been lost in a recording, or to simulate reality to sampled instruments. The instruments in the music appear to have the natural space around them that they would have in reality. This process uses psychoacoustic principles which have been known for years. It is fully mono-compatible and the amount of stereoisation can be controlled to suit any recording.



The Importance of Referencing (Psychoacoustic)Your ears are the most vital tool in the entire process.your ears can suffer fatigue from long periods of listening, and your perception changes depending on the level and frequency of surrounding sounds presentYour ears need a reference, in the way of a piece of well produced professional material



Mastering Summary24 bit data storage and processing. Source recordings and masters should have higher resolution than the eventual release medium.Avoid successive A/D and D/A conversion, as this is a quantisation process which results in a deterioration in sonic quality.Avoid too many processing equipment in the chain.Reduce the practice of “cumulative processing” (such as multiple stages of EQ). Only convert to 16-bit once when all processing and adjustment has been done, and the audio is ready for release.



Dither

When the mastering processes are finished it will be necessary to convert the data back down to 44.1kHz/16-bit so that it can be stored to a master format and duplicated on removable media such as CD. It is apparent that this conversion will introduce some degradation in sound quality for the same reasons as other digital processes. This loss in quality is known as “dither noise” and can appear to “veil” the sound, resulting in a reduction in stereo soundstage, definition and clarity.

Noise Shaping and Dithering Techniques



Super Bit-Mapping

When down-converting to 16-bit from a higher resolution, the extra bits in the word cannot be implemented in the data stream. Super bit-mapping uses the extra bits to increase the accuracy of the least significant bits of the 16-bit samples. High order mathematics are used to do this, resulting not only in the maintenance of compatibility, but also a decrease in noise level

and an increase in sound quality.



The Techniques - Practical steps towards masteringTransferring – The transfer of the audio information to the platform on which it can be processed.Processing – The application of signal processing and adjustment to the audio.Editing – The structural adjustment of the audio.CD Burning – The process of storing the audio on removable media.



ProcessingNoise Reduction (remove hiss, clicks and pops)Clean up noise immediately before or after song start or endApply equalisation to improve tonal balance and clarityAdd compressionAudio enhancementEnhance stereo imaging to give a wider stereo spread



EditingRe-sequence songs into desired order for album, and adjust spacing between eachAdjust individual song levels to ensure consistency throughout the albumIncrease overall output levelFade songs in or out with precisionCreate crossfades between songsEdit song structure, remix songs, (remove intro, swap verses)Transfer between different formats (DAT, CD-R, cassette)Convert sample rates (44.1KHz, 48KHz)



Noise Reduction - Noise can be a problem in all areas of musical recording and production.

Sources of noise can include:Hiss clicks and pops generated by electrical and electronic equipment.Quantizing errors in digital systemsUnwanted noises such as sheet music rustling, microphone spillage from other instruments or other sounds introduced by the environment.



De-clicking

De-clicking is a digital process usually implemented in software, which can remove or reduce clicks or pops in an audio signal which have been caused by digital glitches, or the computers electronics.

De-crackling

Often, mastering engineers are asked to transfer audio from one format to another. In the case of transferring material from a vinyl record to a digital format, it is possible due to the low durability of vinyl that there will be crackling caused by scratches on the record caused by dust picked up in the grooves



Equalisation

Equalisation is a means of controlling the levels of certain frequency areas within a piece of audio. These controls are based around filters, but as well as the ability to “filter out” or “cut” certain frequencies, the engineer is able to “boost” certain frequencies also.



Parametric EQ

In mastering, Parametric EQ is used for pin-pointing and removing unwanted defects at particular frequencies.

Shelving EQ

In other areas of audio production, parametric EQ is the most commonly used type of EQ as it is good for dealing with individual instruments. In mastering however, there is more use for “Shelving EQ”, which is better for whole pieces of stereo material.

Perception of EQ

A person may not be able to hear above 15kHz, they may still hear the effect of equalisation applied above 15kHz.



Perception of Loudness

It is established that the human hearing curve is not flat, but is more sensitive to mid-range frequencies than the extremes of the frequency spectrum.

The Baxandall Curve

Many consumer hi-fi systems have tone controls that are modelled on the Baxandall curve. This curve is applied to boost/cuts in low or high frequency. Unlike shelving EQ, instead of reaching a “shelf” in the curve, the Baxandall continues to rise. This shape is often more pleasing to the ear than any standard shelf.



The Baxandall high frequency boost can be implemented by placing a parametric equaliser (Q=1) at the high frequency limit (20kHz). The portion of the bell curve above 20kHz is ignored,

and the result is a gradual rise from 10kHz to 20kHz.



Application of EQIt is known that the human ear is far less tolerant of EQ boost than of cut, so as little boost as possible will make audio sound more natural.Always use bypass to reference what effect (if any) your application of EQ has made on the material. You must be sure that you really have made an improvement.To add more top end to a mix use a parametric EQ setting with a low Q, to boost at around 12 to 15kHz. This adds “air” to almost any audio which contains high frequency harmonics. Apply broadband boost in the “Air” band.



Below is a list of things to remember when applying EQ to fix a particular instrument within a mix. Take for example a weak guitar solo:Only apply EQ during the solo, there is no need to change the entire song.Only apply EQ to the channel in which the solo is primarily located.Only apply EQ to the fundamental frequencies if possible.



Dynamic Control - Compression at the Mastering Stage

The effect of compression at the mastering stage is to make the music sound “punchier”, louder and more even.

Compression can be used during the recording stage, on individual tracks, to avoid clipping when recording instruments with a wide dynamic range such as voice, or to be used creatively to adjust the “punchiness” of perhaps a drum sound. It is usual to leave the overall compression of the finished stereo song until the mastering stage. There are very good reasons for this.



The mastering house will more than likely have a more advanced compression unit designed especially for compression at this stage, they may use a 24-bit digital compressor with attack, ratio and release times more precise.The mastering engineer will be more experienced about the compromises, advantages and disadvantages of applying overall compression.The mastering engineer must be highly proficient in the control of the compressor, he must be able to program it with precision, adjusting it for the best possible results for each song in question. It is impossible to attempt to make decisions like this at the mixing stage, as you do not have the perspective of the entire album.



The mastering house will monitor the album in the making, using a calibrated monitoring system so that they know exactly how loud it is in comparison with other CD’s of similar music.The mastering house uses non-destructive methods to apply compression so that the engineer can undo anything that is not appropriate or unwanted later on.It is undesirable to apply any kind of overall compression at the mixing stage, before sending to mastering. If the choice of compressor was wrong, resulting in subtle pumping or breathing, or loss of liveliness in the music, the mastering house will have a difficult and sometimes impossible task of undoing the damage.



Limiting - If the compression ratio is very high, for example 10:1, then a situation known as “limiting” occurs. The amount of compression applied to signals above the threshold level is so great, that little or none of these signals get through.

It is important to know when to use limiting as opposed to compression.Consider limiting when you want to increase the apparent loudness of the material without affecting it’s sound.Consider using compression when the material seems to lack punch or strength.



Level Adjustment

In the mastering process, the individual songs for an album are “levelled” whereby all of the songs are brought up to the same volume as each other respectively.



Audio Enhancement

In order to be competitive, producers want the music with which they are working to be larger, and more impressive than life. Using audio enhancement devices, mastering engineers have the ability to:Make the recording sound brighter, more punchy.Enhance the sense of detail, individual sounds are more distinguishable.Make the recordings sound louder, without changing the peak level.



Dynamic Enhancement

Our perception of the audio spectrum changes as sounds become louder. At high listening levels, our ears tend to perceive low and high frequencies as being louder, and at lower listening levels, the mid-frequency range is more dominant.

Input Output

BandpassFilter 6-8kHz

Compressor

+

mix



Harmonic Enhancement

When an audio signal is subjected to intentional or unintentional distortion, high frequency harmonics are produced. These harmonics can be audible, and this is undesirable.

Input Output

Mix Control

Filter frequency control

High passfilter

Harmonics generatorDistortion/compression



Phase Manipulation

It is known that in reality, high frequency sounds travel slightly slower than low frequency sounds. The greater distance from a sound source the listener is, the more the high frequencies will appear to lag behind. Some enhancement devices take advantage of this principle, by delaying the lower and mid-range frequencies, sounds can appear nearer and more immediate.



TipsHarmonic enhancement is best for replacing missing detail in the high end, the likes of which could not be recovered using conventional EQ, and is better than phase manipulation or dynamic enhancement at rescuing dull recordings.Dynamic enhancement is good for inducing a greater sense of detail, and different instruments are more easily distinguishable.Phase manipulation will recover problems due to phase shifts, and also improve detail within the recording.



TipsOnly use as much enhancement as is necessary, as any noise present in the signal will be emphasised also, and this should be kept to a minimum.Make frequent comparisons between the enhanced sound and the original sound to avoid over-enhancement, leading to music which is too bright, and is fatiguing to the listener – the idea is to add air and space to the music, without making it sound unnatural.Improve noisy recordings by following the dynamic noise filter mentioned earlier, by some form of enhancement.



Stereo Enhancement

It is sometimes necessary for mastering engineers to alter the perceived stereo width of a recording. It is possible to create illusions of sounds coming from different points beyond the speakers by using clever processing. There is a simple way of making the speakers sound further apart than they are.Invert the phase of a small part of the right channel and divert it into the left channel. Invert the phase of a small part of the left channel, and feed it into the right channel.



Arrangement of an album

Another job of the mastering engineer is to ensure continuity between all the songs to be included on the album he is working with. They all need to be at a similar level, and of the same quality.



Comparison of your work to a professionally released CD

Conduct a listening test to comparison your work to a professionally released recording. This test can be done through studio monitors, a stereo at home, and a stereo in the car. The aim is to see how the music handles in different speaker situations. Your music must be adjusted accordingly to what sounds good in all of these situations.



Static , Nonstreaming, Downloadable Audio files

Liquid Audio, MP3

WAVE (.wav)

AIFF

AIFF-C (AIFC) - compression format (3:1 & 6:1)

AU (Sun audio, NeXT audio, MU-law, u-Law)

MPEG - lossy compression

QuickTime- Multiplatform

SND (SND Resource) - Hypercard

VOC - Creative Labs



Streaming Audio

Plug-Ins Versus Players

Many players have a corresponding plug-in, and some offer Software Development Kits (SDK) eg. RealAudio by RealSystem G2RealAudio-Custom InterfacesQuickTime Live Webcasting-running on a G4 with solution provided by Sorenson Broadcaster. US$199Http://www.s-vision.com/products/SorensonBroadcaster/index.html

Shockwave:- Compiled from DirectorFlash:- Sound & Animation using Vector Graphics



Tools

Media Cleaner Pro - Power Suite 4.0.2

Dynamic media compression which supports cross-platform, and also supports QuickTime, RealSystem G2, Windows Media, MPEG, MP3 etc. on both MacOS & Windows.

Wed 19/4/2000 Audio Mastering - A Neglected Art Music & Sound for Multimedia & INTERNET.

Documents

audio cd mastering

process of mastering

net2 audio mastering

net1 audio mastering

net3 audio mastering

net4 audio mastering

net5 audio mastering

net6 audio mastering