You are here

Noise Reduction Methods

For everything after the recording stage: hardware/software and how you use it.

Noise Reduction Methods

Postby Straff » Sun Apr 15, 2012 4:39 pm

Hey Guys!

First of all, HI! - New to the SoS forum, however been reading SoS for a while now, just finally got around to joining the forum. Brief intro then i'll get to the point. I'm a 22 year old student studying music & audio production @ Bournemouth university / work p/time as a live engineer/audio tech crew for a music venue in the local area.

Annnnnnndd to my point. Basically i'm researching for a report on the technologies and methods used in various noise reduction methods in audio (both analogue and digital), and am looking for some useful resources of information to base my report around, i've found a few books that are quite useful, but apart from your standard (avoidable if possible) wikipedia and sites like that, i can't seem to find much concrete information on the web. but wondered if any of you vets out there could point me in the right direction for anything on the various approaches to noise reduction methodologies? i.e i've found a website that details the use of dynamic filtering and multi-band expanders to reduce hiss and such.

But wondered if there is anything out there a little more comprehensive? or any practices you experienced engineers/producers swear by/use frequently that I can research?

I've done the library. I've done 'the google', my next process was posting on forums/blogs etc and actually ask some people in the game.

Any help/direction would be massively appreciated!

Thanks,

D.
User avatar
Straff
Poster
Posts: 12
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby Tom Ballinger » Sun Apr 15, 2012 4:45 pm

Hi

the AES (Audio Engineering Society) is very good for books and journals. It cots £20 to join i think you get the annual journal free and then it costs £3 for every book/journal you download. Iv been using it for my degree and i have found it invaluable!

Hope this helps.

Tom.
Tom Ballinger
Poster
Posts: 17
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby dmills » Sun Apr 15, 2012 5:20 pm

Dolby labs website may be informative, as they commercialised a lot of the early work back in the tape era, and don't forget to search their patents for the early stuff, they patented a lot of tape noise reduction methods.

TBH, noise reduction is not the issue it once was as any modern recording chain should be quiet enough to be below the environmental noise floor in 99% of recording spaces (And pretty much all playback spaces).

Regards, Dan.
dmills
Frequent Poster
Posts: 1544
Joined: Thu Aug 24, 2006 11:00 pm
Location: High Wycombe, UK

Audiophiles use phono leads because they are unbalanced people!


Re: Noise Reduction Methods

Postby Hugh Robjohns » Sun Apr 15, 2012 8:06 pm

Noise reduction isn't needed in any digital system running with more than about 14 bit word length at the converters, but was important in the days of analogue tape and analogue radio links --and still is for analogue radio mics.

There are single ended NR systems, mainly involving downwards expansion and/or dynamic filtering, but most good systems are double ended with complementary encode / decode systems.

The simplest is the pre-emphasis / de-emphasis idea, with fixed eq boosting hf on the way in, with the opposite eq on the way out to provide overall flat response, but pushing the medium noise down in the more audible hf region at the same time.

Then there are the compander systems, reducing the dynamic range of the input signal into the medium, and then expanding it back again on the output, pushing all the system noise down along the way. Of course, such systems have tracking issues which may result in audible side effects. The various dbx systems worked along these lines.

Dolbys approach to improve this problem was to split the audio into separate bands and comp and each independently, relying on noise masking within each band to hide the artefacts. Dolby SR took the idea much further than Dolby A, and was very good, but came out as digital took off, so didn't really gain the familiarity it deserved. Dolby b, c and s were simplified domestic systems with less nr efficiency and more artefacts.

dolbys main competitor in Europe was telecom with it's c4 system, worth checking out. Dolbys website has a lot of very good White papers on the technology involved.

It is also worth noting that a lot of the ideas behind effective noise reduction translate to digital data reduction, in the various frequency and temporal noise masking processes involved in hiding the processing artefacts.

Studio Sound magazine had a lot of good articles about this stuff back in the late 1970s and 80s if your library has them.

Hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby feline1 » Tue Apr 17, 2012 9:06 am

The best resource of course is Spinal Tap, which discusses Dubbly Noise Reduction at some length.
User avatar
feline1
Frequent Poster
Posts: 2465
Joined: Sun Jun 22, 2003 11:00 pm
Location: Brighton, UK

~~~ A weasel hath not such a deal of spleen as you are tossed with! www.feline1.co.uk ~~~


Re: Noise Reduction Methods

Postby Hugh Robjohns » Wed Apr 18, 2012 2:10 pm

Ah yes -- I remember them talking about using the Dobley system. It's always fun asking the producer if they want to use a dobley on the recording, and even better if they say yes as they sit down in front of the Pro Tools screen!

Hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby Straff » Wed Apr 18, 2012 2:49 pm


Okay great stuff guys, very helpful all of you! Will check out the concepts/places you have all suggested for more info!

Regarding the point made on noise reduction not being much of an issue in digital systems, is this depdendent on the bit-depth of the ADC's solely? as naturally that would result in larger quantisation errors and therefore Q noise? - or are there other factors within digital signal chains to take into account?

Thankyou all again, your help is very much appreciated!
User avatar
Straff
Poster
Posts: 12
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby Hugh Robjohns » Wed Apr 18, 2012 3:54 pm

straff wrote: Regarding the point made on noise reduction not being much of an issue in digital systems, is this depdendent on the bit-depth of the ADC's solely?

Not solely, but mostly.

...as naturally that would result in larger quantisation errors and therefore Q noise? - or are there other factors within digital signal chains to take into account?

If the digital system has been properly dithered then there will be no quantisation errors at all, but there will be a constant noise floor created by the dither at whatever the wordlength is. So the measured signal-to-noise ratio of a dithered digital system is dependent entirely on the wordlength. The rule of thumb is:

SNR = (6 x wordlength bits)-3dB (the -3dB takes account of the effect of the dither noise)

So a 16 bit system like a CD player has a signal-noise ratio of 93dB, and a 20 bit system will have a SNR of 117dB. Most nominally 24 bit converters struggle to deliver more than about 115dB SNR, although a few high end units might reach the equivalent of true 21 bit performance.

However, the perceived noise floor may be significantly lower if noise-shaped dithering has been employed, since the idea of this is to reduce the noise energy in the regions of the spectrum where our hearing is most sensitive (2-6kHz) and increase it commensurately in the areas where we are less sensitive -- mostly above 15kHz.
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby Straff » Thu Apr 19, 2012 2:11 pm


This is excellent! thankyou for the info, very helpful!

if anyone knows of any good articles or posts on dithering that would also be very handy, but apart from that everything you have all suggested has been great! thanks guys.
User avatar
Straff
Poster
Posts: 12
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby Hugh Robjohns » Thu Apr 19, 2012 2:41 pm

User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby The Red Bladder » Thu Apr 19, 2012 3:08 pm

Hugh Robjohns wrote:dolbys main competitor in Europe was telecom with it's c4 system, worth checking out.


I had this for a while (bought packaged with a Soundcraft multitrack). It sucked! No artefacts, but hardly any NR either! About 3dB, if my memory serves me rightly. Come to think of it, the only effect we were able to hear, was that the audio had the top end taken off, no matter how we tried to set the beastly thing up.

Until SR came along, we used to just do the pre-emphasis thing. It gave us 6-10dB less noise, which, after all, is markedly better than a poke in the eye with a sharp stick!
The Red Bladder
Frequent Poster
Posts: 1880
Joined: Mon Jun 04, 2007 11:00 pm
Location: . . .

 


Re: Noise Reduction Methods

Postby Hugh Robjohns » Thu Apr 19, 2012 3:39 pm

The Red Bladder wrote: I had this for a while (bought packaged with a Soundcraft multitrack). It sucked! No artefacts, but hardly any NR either! About 3dB, if my memory serves me rightly.


I'm surprised. C4 came along well after Dolby A was introduced (in 1965). I think it was launched around 1979 or so, but used similar ideas including splitting the audio into four bands. ANT Telcom improved the level tracking system so that it was much less prone to mistracking if the record or replay levels weren't absolutely spot on -- which was critically important with the Dobly systems.

I believe C4 claimed up to 30dB noise reduction (compared to Dolby A's 10dB risng to 15dB in the high band). Although most of my experience was with Dobly A and later, Dolby SR, I heard a C4 system in use on a 24 track in the early 1980s and was quite impressed with it -- it seemed significantly better than Dolby A, with less audible artefacts and better overall noise reduction.

I'm surprised you found it so disappointing, but at least going down the pre-emphasis/de-emphasis route avoided all the mistracking errors, even if it was at the expense of HF headroom and bias wander. Would have been the ideal solution had Dolby HX been around back then!

Hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby Straff » Thu Apr 19, 2012 6:57 pm

Okay, so from what i've been reading so far, am i correct in saying that the idea behind the pre-emphasis / de-emphasis filtering system is to contribute toward maintaining a constant SNR by filtering frequencies above a cutoff point to increase their amplitude, (normally in the higher end of the spectrum) as the transfer medium of the signal doesn't have a flat frequency response within the bandwidth of the signal's frequency spectrum, causing an attenuation of certain frequencies, and therefore causing things like attenuation distortion.

This whole area of signal processing is still pretty new to me, but i think i'm getting a pretty firm understanding of how things are working?


Keep it coming!

p.s - Thanks for link to your article hugh! working my way through that little gem now!
User avatar
Straff
Poster
Posts: 12
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby The Red Bladder » Thu Apr 19, 2012 7:42 pm

Hugh Robjohns wrote:I'm surprised.


TBH it was a long time ago and it was my first proper job in a pukka commercial recording studio. In other words, it might have been pilot error! But I do remember it sounding dull and getting switched off. The studio PX'ed it for an Otari and after two years I quit to open my own studio and music store. Never worked for the man since! All long hair, jeans, no bloody clue, free beer for the customers and a young wife who sunbathed naked on the studio veranda!

Our present R2R has HX and with pre-de it works a treat - especially with high energy tape and at 15ips.
The Red Bladder
Frequent Poster
Posts: 1880
Joined: Mon Jun 04, 2007 11:00 pm
Location: . . .

 


Re: Noise Reduction Methods

Postby Hugh Robjohns » Fri Apr 20, 2012 10:09 am

straff wrote: ...am i correct in saying that...

The idea of pre-emphasis is two-fold. First, we are generally more sensitive to high frequency noise, and seceond, most musical forms naturally have considerably less energy at the HF end of the spectrum than at the LF end. This means there is a lot more headroom at the high end than the low end.

So, pre-emphasis boosts the high frequencies with a simple EQ curve, taking advantage of the otherwise unused available headroom margin. On replay a corresponding attenuating EQ curve is applied to restore the overall frequency response to a flat line, but in doing so also reduces the HF response of the noise floor by the same amount. This gaining a useful reduction in perceived noise. The EQ is an entirely static process, so there can be no dynamic pumping or other artefacts going on, unlike more sophisticated NR systems like Dobly, dbx etc.

Pre-emphasis / de-emphasis is used in all manner or audio devices: analogue tape recording, FM radio, NICAM TV sound, and was used in the early days of CD and DAT too to improve the poor performance of the first generations of digital converters. Different systems use slightly different curve shapes and turn-over frequencies, but they are all using the same underlying concepts.

hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby Hugh Robjohns » Fri Apr 20, 2012 10:10 am

The Red Bladder wrote:...and a young wife who sunbathed naked on the studio veranda!

Excellent... I must get one of those... can't find anything in the Studiospares catalogue though...

hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby James Perrett » Mon Apr 23, 2012 9:33 am

One thing to remember with a pre-emphasis/de-emphasis system is that you can only go so far before you start saturating the medium. Analogue tape saturates at a lower level for high frequencies than it does for low/mid frequencies. As an extreme example, take a look at a cassette deck frequency response chart and you will see that it is usually measured at -20dB - the 0dB chart would show a big droop at high frequencies.

James.
James Perrett
Moderator
Posts: 5878
Joined: Sun Sep 09, 2001 11:00 pm
Location: The wilds of Hampshire

JRP Music - Audio Mastering and Restoration.
http://www.jrpmusic.co.uk


Re: Noise Reduction Methods

Postby zenguitar » Mon Apr 23, 2012 12:37 pm

Hugh Robjohns wrote:
The Red Bladder wrote:...and a young wife who sunbathed naked on the studio veranda!

Excellent... I must get one of those... can't find anything in the Studiospares catalogue though...

hugh

Gets expensive if you need to upgrade though

Andy
User avatar
zenguitar
Moderator
Posts: 7427
Joined: Thu Dec 05, 2002 12:00 am
Location: Devon

Largely, but not entirely, dysfunctional.


Re: Noise Reduction Methods

Postby ef37a » Mon Apr 23, 2012 1:40 pm

James Perrett wrote:One thing to remember with a pre-emphasis/de-emphasis system is that you can only go so far before you start saturating the medium. Analogue tape saturates at a lower level for high frequencies than it does for low/mid frequencies. As an extreme example, take a look at a cassette deck frequency response chart and you will see that it is usually measured at -20dB - the 0dB chart would show a big droop at high frequencies.

James.

Yea. That's where Dolby HX stepped in.

Dave.
ef37a
Jedi Poster
Posts: 7194
Joined: Sun May 28, 2006 11:00 pm
Location: northampton uk

#They did not listen, they are not listening still...Perhaps they never will?#


Re: Noise Reduction Methods

Postby Hugh Robjohns » Mon Apr 23, 2012 2:05 pm

Indeed it is.

For those unfamiliar with Dolby HX (HX standing for 'Headroom eXtension'), it was a clever idea introduced by Bang & Olufsen in the early 1980s and licensed to Dolby who marketed it as HX-Pro.

Basically, the amount of AC bias in a cassette or tape machine is optimised when the tape machine is aligned to mimimise distortion and noise. However, the high frequency elements in the recorded audio add to that bias level producing an over-bias situation that actually increases distortion and saturation.

By varying the bias level dynamically according to the auadio HF content that over-bias and early saturation can be reduced, effectly reducing HF distortion and increasing the apparent HF headroom quite significantly. This allows much higher recording levels while retaining a linear response.

The system was used in some high-end cassette machines, as well as some of the last generation of pro reel-reel recorders, includng the Studer A807.

hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby ef37a » Mon Apr 23, 2012 2:31 pm

Cheers Hugh,

Did not know B and O were behind HX.
Nice to know they were clever at SOMETHING other than making audio gear that took 5 times as long to take apart and put together than anyone elses!

One could just about forgive this if their hi fi was of the best grade but it was decidedly of the Continental "Boom& Tish" stamp! I think tho' that they did produce a very unusual speaker design not that long ago with a superb performance?

Dave.
ef37a
Jedi Poster
Posts: 7194
Joined: Sun May 28, 2006 11:00 pm
Location: northampton uk

#They did not listen, they are not listening still...Perhaps they never will?#


Re: Noise Reduction Methods

Postby Hugh Robjohns » Mon Apr 23, 2012 4:00 pm

B&O had some very clever people and came up with several advanced ideas... but like you I was never really impressed with the 'lifestyle' implementation or the general sound quality.

hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby ef37a » Mon Apr 23, 2012 4:14 pm

Hugh Robjohns wrote:B&O had some very clever people and came up with several advanced ideas... but like you I was never really impressed with the 'lifestyle' implementation or the general sound quality.

hugh

Yes! Their colour TVs in particular were a miracle of compact, advanced design but a total illegitimate to service!
Fortunately I never worked for a B&0 dealer but I did do 10 years for a Grundig one, almost as bad! WTF use high speed SCRs in a line timebase?! Everyone else was fine with transistors!(power FETs I guess in the last of the CRTs? Bet they never failed?)

Dave.
ef37a
Jedi Poster
Posts: 7194
Joined: Sun May 28, 2006 11:00 pm
Location: northampton uk

#They did not listen, they are not listening still...Perhaps they never will?#


Re: Noise Reduction Methods

Postby Straff » Wed Apr 25, 2012 12:35 pm


The Dolby preemphasis boosts the recorded level of the quieter audio signal at these higher frequencies during recording, effectively compressing the dynamic range of that portion of the signal, so that quieter sounds above 1 kHz receive a proportionally greater boost. As the tape is recorded, the relative amplitude of the signal above 1 kHz is used to determine how much pre-emphasis to apply - a low-level signal is boosted by 10 dB (Dolby B) or 20 dB (Dolby C). As the signal rises in amplitude, less and less pre-emphasis is applied until at the "Dolby level" (+3 VU), no signal modification is performed.


Just getting to grips with the various Dolby NR systems and how they operate, but this particular section i've just read doesn't make complete sense to me?

How does boosting frequencies above a cutoff point (i.e 1kHz) effectively compress a signals dynamic range? surely you are making use of the DR and headroom available to emphasise those frequencies? in then goes on to say that the rise in those HF signal amplitudes is THEN used to determine how much pre-emphasis to use?

i'm not sure if i've missed something when trying to understand that excerpt, but surely creating a boost in the frequencies where tape hiss is more audible IS the actual pre-emphasis part of the system at work?

User avatar
Straff
Poster
Posts: 12
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby The Red Bladder » Wed Apr 25, 2012 12:44 pm

You are confusing variable and non-variable.

Dolby systems are variable - more highs, less boost.

The pre-de thing done on the desk with totally non-variable and every engineer had their own way of doing this. As long as they wrote down on the tin, what they had done, all was hunky-dory.
The Red Bladder
Frequent Poster
Posts: 1880
Joined: Mon Jun 04, 2007 11:00 pm
Location: . . .

 


Re: Noise Reduction Methods

Postby Straff » Wed Apr 25, 2012 12:53 pm

The Red Bladder wrote:You are confusing variable and non-variable.

Dolby systems are variable - more highs, less boost.

The pre-de thing done on the desk with totally non-variable and every engineer had their own way of doing this. As long as they wrote down on the tin, what they had done, all was hunky-dory.

RB, I apologise in advance - but that blew my mind a little bit, i'm not sure I followed that at all? - feel free to elaborate, it will eventually stick.
User avatar
Straff
Poster
Posts: 12
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby Exalted Wombat » Wed Apr 25, 2012 1:29 pm

Although NR should never be required in a recording project today, you need to know how to deal with older tapes recorded with Dolby etc.

The technology which isolates unwanted noise from wanted signal is still very relevant in attempting to pick out one voice from a crowd. The fact that military resources must be pouring into this problem is our best hope for spin-off technology that will "un-bake the cake" of a full mix down to its individual instruments.

Dithering is an interesting subject. Although it's easy to design an experiment to show its effect, in much real-life mixing it can be hard to justify.
Exalted Wombat
Jedi Poster
Posts: 5652
Joined: Sat Feb 06, 2010 12:00 am
Location: London UK

You don't have to write songs. The world doesn't want you to write songs. It would probably prefer it if you didn't. So write songs if you want to. Otherwise, dont. Go fishing instead.


Re: Noise Reduction Methods

Postby Hugh Robjohns » Wed Apr 25, 2012 2:41 pm

Straff wrote:Just getting to grips with the various Dolby NR systems and how they operate, but this particular section i've just read doesn't make complete sense to me?


I'm not surprised, it's a pretty convoluted explanation.

Basically what it is saying is that the audio spectrum is divided, and high frequencies are compressed to ensure they are well above the noise floor, the filter turnover is varied according to the spectral content of the programme material. On playback the opposite expansion is applied to restore a uniform overall frequency response. Dolby B applied relatively little compression, Dolby C applies rather more (which is also why it is much more sensitive to compressor mistracking), while Dolby A splits and processes four fixed bands and Dolby SR five (but in a more complicated way with both fixed and variable bands!).

You'd be better off reading Dolby's own papers for understandable explanations:

http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/1000_kens_corner.pdf

http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/212_Dolby_B,_C_and_S_Noise_Reduction_Systems.pdf

http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/151_363_8.Manual.pdf

This last one is the user manual for a Dolby A/SR codec, but includes info on the working principles in section 5, and a whole lot more in several papers at the end.

hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Noise Reduction Methods

Postby Straff » Thu Apr 26, 2012 9:33 am

Thanks Hugh, getting through those now, However;

in the first of those dolby documents, it words the process of increasing the amplitude of the HF spectrum similarly as:

To avoid losing the quiet music in the hiss, we could turn up the recording level. However, we can't just leave it turned up because the next loud passage that comes along will go hard "into the red", causing gross distortion in our recording. In principle, we could "ride the gain", that is, continually turn the level up when the music is quiet and down when it is loud. However, the composer and performers presumably intended the soft passages to be soft, and would most probably object to our making them louder than they should be, that is, compressing the music's dynamic range.


Now i think i've been picturing this in the wrong light, as with the previous quote. When they both mention effectively compressing the song/audio's dynamic range from increasing the HF amplitude, i'm thinking that the actual signal being accentuated can't possibly 'compress' the dynamic range because you are in fact expanding that signals varying amplitude range?

However, am i correcting myself appropriately when i say that they are both not speaking of the specific frequency bandwidth being amplified, but as of the whole composition/track being 'compressed' as the dynamic variety of the song's high/low level sections is less varied/dramatic? - i think this is what they are intended to get across, but i first interpreted it as the above explanation.
User avatar
Straff
Poster
Posts: 12
Joined: Sat Apr 14, 2012 11:00 pm

Re: Noise Reduction Methods

Postby Hugh Robjohns » Thu Apr 26, 2012 9:59 am

Straff wrote:Now i think i've been picturing this in the wrong light, as with the previous quote.

Your quoted section is not the complete story, and you're taking it out of context.

Yes, raising or 'compressing' the quiet signals (especially high frquency quiet signals) to ensure they are well above the noise floor is an important part of the noise-reduction process... but as your quote says, to do only that would be to destroy the original musical dynamics.

So a complete noise reduction system has to 'undo' or restore these dynanmic changes on replay, using 'expansion'. That's why systems like dolby are called double-ended noise reduction and employ complementary encode and decode processes.

i'm thinking that the actual signal being accentuated can't possibly 'compress' the dynamic range because you are in fact expanding that signals varying amplitude range?

No. If the original tracks loudest peak was, say at +10dB and the quietest was at -50dB, we have a total dynamic range of 60dB. If the quietest bits are raised to -20dB to ensure thay are well above the noise floor, the dynamic range has been reduced or 'compressed' to just 30dB -- half as much, or a 2:1 compression ratio.

The decode process will restore the original dynamic range by 'expanding' it back to the full 60dB with an expansion ratio of 1:2.

However, am i correcting myself appropriately when i say that they are both not speaking of the specific frequency bandwidth being amplified, but as of the whole composition/track being 'compressed' as the dynamic variety of the song's high/low level sections is less varied/dramatic? - i think this is what they are intended to get across, but i first interpreted it as the above explanation.

Some systems, like DBX, work with a single compress/expand process across the entire audio spectrum in one go. This is easy to do, but the down side is that the dynamic changes are not masked very well and can be heard as unwanted artefacts.

Dolby B and C only process the high frequencies, where noise masking is more effective. Dolby A, SR and S split the audio into more separate frequency bands and process them indpendently to take greater advantage of noise masking in the ear's critical bands.

Hugh
User avatar
Hugh Robjohns
Moderator
Posts: 17082
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Next

Who is online

Users browsing this forum: No registered users and 2 guests