H

Hazvinei

guest
Novel online mastering services like LANDR, ARIA, eMastered, etc are they a threat to mastering engineers' careers? If so explain why robot mastering isn't particularly a good idea to be widely implemented?
 
recommended
@Munyaradzi Mafaro unfortunately that's true. If that wasn't true there wouldn't be loudness wars in the music industry. All you have to do is just listen to some modern Jazz songs and decide for yourself. But still and all, people are free to do whatever they want with their music so that's that.

@Tafadzwa Twabam software companies like iZotope, FabFilter, Waves Audio, DMG, etc are also a threat to professional mixing and mastering engineers because they have been progressively making it possible for each and every bedroom producer to easily have access to state-of-the-art software without the need to spend thousands of Euros/Dollars like in the old days, neither do you need to be someone's tea boy/girl to learn the craft...

Chidinma

Moderator
Machine-learning "cloud mastering" companies have bigger marketing budgets, so they pose a serious threat to emerging mastering engineers who can't afford to digitally market themselves properly and cut through the noise e.g. paid search on Google & Bing including other notable music publications & content creators e.g. YouTubers with a huge pro audio following.

But right now they aren't much of a threat to established mastering engineers who have already built strong relationships with both major and independent record labels—they are gonna be fine for a while.

In short, the writing is on the wall for those capable of interpreting it properly.
 
Audio mastering in modern times has been watered down to two things and these are:
  • Brighter is ALWAYS better.
  • Louder is ALWAYS better.
And with that type of mindset, the moral of the story is ML (i.e. machine learning) can go beyond human intelligence and can be taught to do this efficiently hence automated mastering will assure consistency and reproducibility as the algorithms become more and more advanced.

Eventually, mastering engineers will be put out of business by such services followed by mixing engineers (i.e. automated audio mixing with minimal or no human involvement at all), and of course, machine learning will come for music producers, songwriters then as the final nail on the coffin it will come for record label employees.

When it's all said and done it's the investors who stand to benefit a LOT, I mean a LOT than any other period in music business history.

Why? Because human beings are so expensive and less productive—you can't in most places legally get them to grind 100 hours a week (even if you allow them to piss in bottles) compared to a machine/program that doesn't need to take breaks or ask to go on holidays neither does it expect a bonus and all the other work-related benefits! All it needs are regularly scheduled updates—nothing more and nothing less.

N.B. companies don't exist primarily for the purpose of providing jobs. Their purpose is to provide a free market with a product or service. In other words, jobs help them reach their goals although this isn't a priority.

I guess this is arguably why some billionaires constantly advocate for population control—they sure know what's coming (i.e. perhaps 500 million people on earth) since if current jobs/careers become widely automated what then will happen to billions of people except chaos and unending daily protests?

In other words, I suppose to them a reduction of population is much more manageable on universal basic income than billions of people doing nothing but bonking each other every week whilst popping more babies year after year.

The offspring won't find work when they become adults, still.
 

Sadzandiuraye

initiate
By the way, are we talking specifically about mastering music? Because mastering can also be done to podcasts, YouTube videos, etc.

Otherwise, the work of human mastering engineers is mostly based on trial and error, as well as personal prejudice (biased preferential tastes about what something should sound like) this is why if you give 10 different audio mastering engineers to work on a song you get all kinds of results.

And although automated online mastering is not perfect, it can yield incredible results for an independent musician who can't afford to pay a human mastering engineer to work on their music.

I personally look at it as an alternative solution and there is nothing wrong with introducing alternative solutions.

The problem though there is a potential of subversive propaganda—I mean aggressive public relations i.e. op-eds on re-educating the masses about what audio mastering entails so that these online services can gain a huge market share into the music industry simply by redefining what audio mastering is.

Of course, the target group is always the young generation because you ought to catch them whilst their young. Old heads are already stuck in their ways so much that it's not beneficial to try to subvert them since some of them may literary be one foot into their graves.
 

Scorpio

grand master
What about iZotope's Ozone Master Assistant?

iZotope Ozone Master Assistant is currently meant for people who already have a little idea of how to master a song from scratch. It's different from something like LANDR that is marketed for people who have absolutely no clue what to do with a mix.

But make no mistake about it, this is going to change as major updates come along in years to come. And in those days Ozone Master Assistant will receive just as much hate as other robot mastering services from audio professionals when it becomes more and more ubiquitous with the definition of "audio mastering" like what @Sadzandiuraye pointed out.

However, one thing I have noticed about Ozone Master Assistant is that it always seems to set the Maximizer's character to 0.6 which distorts like crazy. You can fine-tune it though which is a good thing. I don't think you can currently do that with all these other automatic mastering services on the web.
 

Lenard

Lenard Maseko
@Munyaradzi Mafaro unfortunately that's true. If that wasn't true there wouldn't be loudness wars in the music industry. All you have to do is just listen to some modern Jazz songs and decide for yourself. But still and all, people are free to do whatever they want with their music so that's that.

@Tafadzwa Twabam software companies like iZotope, FabFilter, Waves Audio, DMG, etc are also a threat to professional mixing and mastering engineers because they have been progressively making it possible for each and every bedroom producer to easily have access to state-of-the-art software without the need to spend thousands of Euros/Dollars like in the old days, neither do you need to be someone's tea boy/girl to learn the craft. You can do it on your own through patience, perseverance coupled with trial and error.

Now think about it? Why pay something like $1200 to get an album mastered when you can try each and every online mastering service or program with that type of budget and decide what you like and what you don't like. This is where the industry is going—a race to the bottom just like on Amazon Seller Central with all those constant repricing shenanigans by pennies.

Furthermore, automated mastering in the event that it vastly improves by a long shot it is actually good for major record labels who regularly have to release music each and every week. Imagine the huge amounts of money they could save or reinvest into a business instead of paying several human mastering engineers thousands of dollars on music projects.

It's just business and if you can't compete you are straight out of luck. Unfortunately, that's how this world works. It's either you beat them or you join them— it's all part of the great capitalist game.
 
recommended

Queen

grand master
I just wanna know how does a cloud audio mastering service or software fully understands why this mix, in particular, sounds really, really bad and needs to be fixed ASAP instead of wasting time trying to do something about it?

What if you feed it with an already mastered track from yet another online audio mastering service what does it do? Add more distortion on top and ask you if you have any comments before you proceed to download your final WAV file?
 

Tanonoka

initiate
@Queen you have already answered your second question because that's exactly what happens when you upload an already mastered song.

Most algorithm-based audio mastering services as we speak are meant for stuff that's hasn't yet been mastered (or content without too much stereo bus processing) otherwise the results will sound worse than a bedroom music producer fooling around with iZotope Ozone, Brainworx bx_masterdesk, or IK Multimedia Lurssen Mastering Console. Hence, there are NOT that "intelligent" as they are marketed in the press.

I'd suggest that at least these services should check the average or integrated LUFS/LKFS of the program material. If the integrated LUFS/LKFS is between -8 to -10 LUFS/LKFS obviously it means someone intentionally slammed a mix into a brickwall limiter so there's no point in doing anything to that except to recommend that they should re-upload an unmastered version of the same mix.

Finally to answer your first question, no, most of these robot mastering services currently aren't meant to give you further feedback about your mix neither what they did to make it sound better. They are simply meant to process audio based on whatever proprietary algorithm the platform thinks is suitable to implement for a given style of music.
 

Queen

grand master
@Tanonoka so basically they automatically pick pre-defined presets to apply to a mix? For example, if there is too much bass at 50 Hz do this, and if there is less bass do that. Once you are done compare that to other popular material of the same genre until there isn't further processing needed?
 

Tanonoka

initiate
@Tanonoka so basically they automatically pick pre-defined presets to apply to a mix? For example, if there is too much bass at 50 Hz do this, and if there is less bass do that. Once you are done compare that to other popular material of the same genre until there isn't further processing needed?

Well, yeah that's a simple version to understand this.
 

Lenard

Lenard Maseko
@Tanonoka you may be right in what you are saying.

But I am really biased to think that ALL robot mastering services have the capability and can literarily detect the average loudness of a song—I mean honestly speaking that shouldn't be a hard task to automate at all.

However, because of capitalist greed, they may be programmed not to discard any audio processing so that customers can still get billed and receive their WAV files even when they intentionally upload already mastered songs just to test if the robot is in actual fact "intelligent" or "retarded."
 

Similar threads

T
Replies
8
Views
195
Minimum Reputation
M
E
Replies
7
Views
83
Dzvuke
D

New Posts

Latest resources

Top