Technical Advacnces in DIY Location Sound

I mean the audio equivalent of this super high res shot (scroll in to any of the pin heads in the crowd for a perfectly clear close-up)

http://gigapan.com/gigapans/033ef14483ee899496648c2b4b06233c?fullscreen_redirect=true

Something that captures every audio detail in a way that allows you to isolate them and use them without fuzz.

In short, no!

Digital audio recording technology has already evolved past the equivalent of the "super high res shot" or at least, well beyond the absolute maximum limits of the human ear and in some cases has evolved up to the limits of the laws of physics. The weaknesses in the audio chain are now mainly just in the transducers (mics and speakers) which convert sound waves to/from electrical energy. The basic technology of transducers (and amplifiers) has not changed greatly in well over half a century, essentially it's just been tweaked over the years rather than having been completely revolutionised, as has happened with analogue audio recording/editing or filming/editing on 35mm or video tape for example. There is nothing on the horizon of which I'm aware which is likely to change this situation.

The most likely advance is in the area of audio software processing: The ability of software to identify and remove noise and other unwanted audio content from existing recordings. The difference between what's wanted and what's unwanted is a human perception though, something computers are not good at and therefore progress is slow and imperfect.

Psychoacoustics is the science of how we perceive sound and there are still big gaps in our understanding, so even measuring many of our perceptions of sound is not yet possible with any great accuracy, let alone actually developing software to process it. So for the foreseeable future, by far the best technology for recording, designing and mixing sound is still a human being!

G

Ah. Great info. So any news on audio software advancements?

Indiefilming said:

Ah. Great info. So any news on audio software advancements?

Click to expand...

Well, there are slow but steady improvements in noise reduction and dialogue manipulation tools. For example, there is a new tool to help remove acoustic reflections (reverb) from production sound, which is more effective/realistic than previous methods. There are also advances in loudness measuring and control being introduced in both the USA and Europe.

G

As the post guys said, filters (generally attached to very expensive software) are improving all the time, but they can't work miracles.

Isolating one audio wave from another audio wave is going to take some very advanced physics, and can I see it being far more complicated than just isolating one color value from another. I'm not sure how you'd pull it off without some high tech sorcery, which would be seriously outside the ability of the average DIYer.

I know exactly what you want... you want great production sound without having the "big" audio crew - production sound mixer, boom-op and audio wrangler. It is very easily accomplished; you just have to spend huge amounts of money and expand huge amounts of time and exquisitely plan each shot.

Oh, wait, you don't want to spend any money, you don't want to put in the extra effort, and you're shooting run & gun.



Never mind....

ClockworkNinja said:

Isolating one audio wave from another audio wave is going to take some very advanced physics, and can I see it being far more complicated than just isolating one color value from another. I'm not sure how you'd pull it off without some high tech sorcery, which would be seriously outside the ability of the average DIYer.

Click to expand...

To be honest, what you describe would be relatively simple to achieve. Unfortunately the problem is far more complex with sound than it is with colour because all naturally occurring sounds are made up of whole series of of audio waves (the fundamental and harmonics) which we then perceive as a single sound. The sound of a door closing for example contains a fundamental and anything from 10 or so harmonics to dozens but to make it more complicated a door closing isn't a single sound but a whole series of different sounds, each part of the door hits different parts of the frame, the slight whoosh of the air, the catch making contact, etc. Each one of these has a fundamental and a series of harmonics, so there's likely to be hundreds of different frequencies (sound waves) in the sound of a door closing! Now let's say we have some dialogue at the same time as the door closes, the dialogue too is a whole bunch of sounds each with it's own fundamental and harmonic series. The problem we have in separating the door from the dialogue is trying to find out which frequencies belong to which sound and it's even worse than that because the chances are that both the door closing and the dialogue would contain many frequencies in common. For example we are likely to find that both sounds contain harmonics at 2kHz, we can measure how much 2kHz is in contained in the sound but there's no way of knowing how much of that 2kHz belongs to the door and how much of it to the dialogue.

Our ears (our perception of sound) work quite differently to our eyes. I'm not an expert on colour but I'll try to draw an analogy. Let's say we take a shade of red and a shade of pink and mix them together, what we are going to see is a new deeper shade of pink and if you gave this new shade of pink to someone else or to a computer, there would be no way of analysing this new shade of pink to find out what our original shades of pink and red were. The ear doesn't work this way, it somehow does not "see" the new shade of pink, instead it "sees" the original shades of pink and red at the same time. This analogy is much the same as the door closing sound occurring at the same time at the dialogue problem, we don't hear some new sound, we still hear both the door and the dialogue as separate sounds, just occurring at the same time. There are various theories about how the ear (or rather the brain) does this but no one knows for sure.

G

Alcove Audio said:

I know exactly what you want... you want great production sound without having the "big" audio crew - production sound mixer, boom-op and audio wrangler. It is very easily accomplished; you just have to spend huge amounts of money and expand huge amounts of time and exquisitely plan each shot.

Oh, wait, you don't want to spend any money, you don't want to put in the extra effort, and you're shooting run & gun.



Never mind....

Click to expand...

Whoooa down boy. What independent filmmakers want and what resources allow for are two very different things.

I don't know any lazy independent filmmakers who are short on effort, or who hesitates to throw every last dime at their movie.

I've also never met sound people who love their job so much they're willing to work on a project simply for the artistry of it, for little or no money, like the filmmaker counterparts (writers, directors, DPs, Actors etc). This is why many of us would appreciate a technological advancement in this area. Of course, audio people will always have work, as have lighting technicians, electricians, camera operators, grips and best boys since the advent of inexpensive, low-light cameras, and the great movies that have been made with these cameras and little to no crew.

Jean Cocteau: "“Film will only become an art when its materials are as inexpensive as pencil and paper.”

AudioPostExpert said:

Our ears (our perception of sound) work quite differently to our eyes. I'm not an expert on colour but I'll try to draw an analogy. Let's say we take a shade of red and a shade of pink and mix them together, what we are going to see is a new deeper shade of pink and if you gave this new shade of pink to someone else or to a computer, there would be no way of analysing this new shade of pink to find out what our original shades of pink and red were. The ear doesn't work this way, it somehow does not "see" the new shade of pink, instead it "sees" the original shades of pink and red at the same time. This analogy is much the same as the door closing sound occurring at the same time at the dialogue problem, we don't hear some new sound, we still hear both the door and the dialogue as separate sounds, just occurring at the same time. There are various theories about how the ear (or rather the brain) does this but no one knows for sure.

G

Click to expand...

Fantastic analogy.

Indiefilming said:

I don't know any lazy independent filmmakers who are short on effort, or who hesitates to throw every last dime at their movie.

Click to expand...

Really, I've met quite a few.

Indiefilming said:

I've also never met sound people who love their job so much they're willing to work on a project simply for the artistry of it, for little or no money, like the filmmaker counterparts (writers, directors, DPs, Actors etc).

Click to expand...

I've met many hundreds of sound people over the couple of decades I've been in the business and many hundreds more sound students, apprentices, etc. I don't know of a single one of them who hasn't worked for little or no money!

Indiefilming said:

This is why many of us would appreciate a technological advancement in this area. Of course, audio people will always have work ...

Click to expand...

These two sentences are mutually exclusive. If there were a way to record perfect sound on set without spending money on a production sound mixer, why would anyone ever employ a production sound mixer?

Indiefilming said:

Jean Cocteau: "“Film will only become an art when its materials are as inexpensive as pencil and paper.”

Click to expand...

So Jean Cocteau believes no film ever made is art, including his own. I wonder why he bothered making any films then if they weren't art, what was he trying to do I wonder?

G

AudioPostExpert said:

Really, I've met quite a few. G

Click to expand...

Hmm. I live in Los Angeles and have worked in this business a long time and maybe I'm just super lucky that all the indie filmmakers I know are quite productive and resourceful. Without exception.

AudioPostExpert said:

I've met many hundreds of sound people over the couple of decades I've been in the business and many hundreds more sound students, apprentices, etc. I don't know of a single one of them who hasn't worked for little or no money! G

Click to expand...

If you know any who are really good please let me know. I'm not kidding. I have not found this to be the case. At least with anyone who can record sound better than I can (and I am not qualified). If you know any who are really good and want to work on cool projects for little money, please PM me.

AudioPostExpert said:

These two sentences are mutually exclusive. If there were a way to record perfect sound on set without spending money on a production sound mixer, why would anyone ever employ a production sound mixer?G

Click to expand...

I don't think in real life these sentence are mutually exclusive. My point was my own real-life observation that lighting technicians, electricians, best boys etc have not been rendered obsolete even though technology exists that has allowed for great film making without them (if the filmmaker is creative and resourceful enough). These people are still being employed by productions that can afford to employ them. Is it harder without them, yes, but independent filmmakers have to work that much harder and be that much more resourceful because they cannot afford a whole crew.

AudioPostExpert said:

So Jean Cocteau believes no film ever made is art, including his own. I wonder why he bothered making any films then if they weren't art, what was he trying to do I wonder? G

Click to expand...

If I may take a stab at it, I believe the point Cocteau was making in this famous quote was that the playing field has to be leveled in order to be viewed as creatively equal to mediums like painting, music, dance, etc. So far, films consider art may have been produced, but for the entire medium to be considered an art form for Cocteau, the exclusivity and elitism due to prohibitive cost structures has to be reduced or eliminated. I believe that is what he meant. Many great filmmakers will never be realized because of so many costly barriers. It's an interesting perspective from someone who made great films.

G[/QUOTE]

Indiefilming said:

If you know any who are really good please let me know. I'm not kidding. I have not found this to be the case. At least with anyone who can record sound better than I can (and I am not qualified). If you know any who are really good and want to work on cool projects for little money, please PM me.

Click to expand...

I don't know so many production sound mixers (PSMs) and those I do know are mainly in the UK. On the audio post side, my business for example is specifically setup for low budget indy film sector but I don't want to turn this into an advert so I'll leave it there.

It also depends on what you mean by "little money", professional PSMs have spent many thousands on equipment and have rent and other living expenses to cover, so working for nothing or zero is simply not an option. Audio Post is even more of a problem because a decent audio post facility/person will have spent many tens or hundreds of thousands on construction/equipment and in addition to personal living expenses there are also the operating costs of the facility to cover. So I would have to charge something in order to be working for free otherwise, if I charged nothing, it would actually be costing me to do the gig. As with composers though, PSMs and audio post guys studying or starting out in the business usually have to work for little or nothing quite a bit before they can start charging and when they do start charging they either have to take out loans, plough almost everything they earn into equipment or both.

Also, I think your basic premise is unfair. It's true that the vast majority of Indy Filmmakers work for nothing and sink their own money into a project for the love and privilege of realising their vision. The audio personnel on the other hand are ultimately working to realise someone else's vision. If I designed a film around and based upon sound design, where everyone was working exclusively towards realising my vision and I had ultimate authority over all the creative content then I would expect to invest heavily for that privilege.

Indiefilming said:

I don't think in real life these sentence are mutually exclusive. My point was my own real-life observation that lighting technicians, electricians, best boys etc have not been rendered obsolete even though technology exists that has allowed for great film making without them (if the filmmaker is creative and resourceful enough).

Click to expand...

The film industry is replete with examples of jobs lost to technology. The jobs may not be completely extinct but the numbers required to do the jobs have drastically reduced.

Indiefilming said:

If I may take a stab at it, I believe the point Cocteau was making in this famous quote was that the playing field has to be leveled in order to be viewed as creatively equal to mediums like painting, music, dance, etc. ...Many great filmmakers will never be realized because of so many costly barriers. It's an interesting perspective from someone who made great films.

Click to expand...

Although Cocteau is legendary, that doesn't mean that everything which ever came out of his mouth is genius. In this instance I disagree with him entirely because almost all art is exclusive and elitist. How many great authors, musicians or painters have not been realised because of religious, political or economic circumstances? When you're destitute and dying of starvation in a drought in Ethiopia, if you get a tiny bit of money, I guarantee a canvas and set of acrylic paints (or even paper and pencil) is not going to be anywhere near the top of your shopping list. Neither is a concert violin or music, ballet or art tuition. What about the societies and religions do not permit women to express themselves artistically. I would say that the vast majority of the world's potentially great artists (in any field) have probably not been realised for economic (or other) reasons and so Cocteau's statement is itself exclusive and elitist. That's just my opinion though!

G

technology improves on an exponential curve, not a linear one. Twice the compute capacity every 2 years, do the math.. in less than 30 years you will be able to buy enough compute capacity for $1000 that you will be able to model every neuron in your brain.. dong the kinda of action you describe will be trivial. So yes.. just not yet..

wheatgrinder said:

...Twice the compute capacity every 2 years, do the math.. in less than 30 years you will be able to buy enough compute capacity for $1000 that you will be able to model every neuron in your brain.. dong the kinda of action you describe will be trivial. So yes.. just not yet..

Click to expand...

Interesting you say "do the math" because that's the heart of the problem! The problem isn't computing power, the number of mathematical equations which can be computed per second, the problem is in defining what mathematical equations need to be computed in the first place. Much of how we perceive sound simply isn't known or well understood yet. Once the scientists have that figured out, another genius has to come along who can translate that knowledge/understanding into a set of mathematical equations and then the final stage is for a computer programmer to create software which uses those equations to solve/automate the tasks we want to accomplish. None of these steps have anything to do with computing power. If we had the understanding, the math and the software, the chances are that the technology and computing power we have today would already be more than sufficient.

To put the problem into perspective, at the moment we don't even fully understand something as simple and basic as how we perceive the pitch of a single musical note and we've been studying it for over a century! So will we be able to do what the OP is asking in the future ... Possibly ... but not in the foreseeable future.

G

your projecting the future linearly, that is a common mistake. Technology is exponential in its rate of change. We are on the knee of the exponential curve. Not only will compute POWER increase exponentially, but so will compute speed.

What the OP is asking is essentially a pattern recognition problem. We, as humans with our slow brains and limited ears can do exactly what the OP wants, we do it every day in casual conversation, talking in restaurants etc.

You underestimate the computational ability of the human brain if you think this is possible to do now. We are still several orders of magnitude away from that power. But like I said, the exponential nature of technology growth will have us OVER That hump in 30 years or so.

You also have to consider the exponential improvements in medical imaging. same thing.. when we can image the state of each neuron in the brain in REAL time, combined with a computational substrate to run it on, we will essentially be able to model the incredibly complex cognitive work our brains do to untangle casual conversation into discrete streams of dialogue.

Uhoh, there I go talking Crazy again..

I'd like to take issue with a couple of points you've made:

wheatgrinder said:

What the OP is asking is essentially a pattern recognition problem.

Click to expand...

Not really, at some point in the future it might become a pattern recognition problem but there are some huge advances in our understanding required before we get to that stage. Before we can recognise a pattern we have to be able to define that there is a pattern and at the moment we can't. There are examples of sounds which are identical in every respect, using every known method of measurement and analysis but which sound different. How do you recognise different patterns in 2 absolutely identical sound files? A computer can only do what it's told and if we don't know what to tell it, a trillion times more computing power or technology is not going to make a jot of difference.

wheatgrinder said:

Not only will compute POWER increase exponentially, but so will compute speed.

Click to expand...

The facts don't agree with this statement. Computing power has increased along the lines stated in Moore's Law but computing speed hasn't changed for many years and this won't change until we either completely change the basis of computing technology or change the laws of physics! Computing power has increased by breaking down tasks into smaller, more numerous tasks and then executing them concurrently using multiple processing threads and cores. So for example, when rendering some video, you could split the video in half, give each half to a different core to process and recombine the two halves when you're done, thereby getting the job done in roughly half the time it would take a single core. But, the actual speed at which an individual CPU core can execute instructions is no faster today than it was a decade ago. This may just be a semantic difference for video processing but not so for audio, as a lot of digital audio processing is recursive. Meaning that many audio tasks can't be broken down into smaller tasks because task B requires the results from task A, so tasks A and B cannot be performed concurrently in different threads or on different cores. That's why for audio processing, fewer cores at a higher clock speed out performs more cores at slower clock speeds. A good example of all this is the 192kHz sampling rate which cannot be implemented accurately due to a processing limitation first identified well over a decade ago. No progress has been made with this flaw after all these years because although processing power has increased dramatically the processing limitation causing the flaw is essentially the speed at which electrons move through an electronic circuit. To cure this flaw requires either a change in the laws of physics or some new form of computing which isn't based on electons flowing through an electonic circuit.

G

I had a 2MHz 8088...
Now I have a 3.2 GHZ i7

Seems like an increase in SPEED to me?

wheatgrinder said:

Now I have a 3.2 GHZ i7

Click to expand...

The pentium 4 had a 3.2GHZ processor nearly a decade ago and over clocking processors to this speed was possible earlier than that.

Maybe I'm confused, would this be a linear or exponential increase in speed?

G