Pitch correction - individual notes within chords

this came up on another thread a while back ... looks almost too good to be true. if it's for real, i'm sure we'll be hearing about it in the trade publications soon.

 

It definitely works- it isn't the easiest app in the world to use, but this is going to change things.... I don't know if it is better or worse, but it is definitely changin'.

 

This is the culmination of ten years of research in the field of Music Information Retrieval. Not only will it help correct tunings and mistakes, but it can search music files for patterns, such as motives, licks, melodies, etc. This is a ground-breaking development.

 

I should mention that software related to the one mentioned here can identify instruments better than humans can. Right now this means oboe vs. soprano sax and the like. Certainly Tele vs. Les Paul can be identified, depending on effects, amps, and recording techniques. The research suggests that if humans can distinguish the difference, then computers can too, but better. Better than a group of listeners. I'm sure a trained expert can outperform a computer at this level, but who knows for how long? Again, there are many blurring factors to take into account.

 

This looks huge. I saw their video and to paraphrase: "What could not be achived theroetically can now be achived practically"

To oversimplify, is this all the result of complex recognition algorithms?

 

if this is where the music industry is going, it's a Pandora's box -- pop music records will sound even more "perfect"! but what would Robert Johnson do?

i've thought for a long time that the ever-more-perfect sound of pop is a factor in the fall-off in live music audiences -- ewwww, that sounds too human!

 

No expert, here, but I don't see why this would have ever been thought to be "theoretically impossible". Simple spectral analysis will let you know what individual pitches (and overtones) comprise a chord. The mathematical tools for moving back and forth between the time and frequency domains have been around for a couple of hundred years.

Maybe it was tricky to sort out desirable frequency components that are are part of a note, but not one if its obvious harmonic frequencies?

Anyway, I believe that:

1) it can be done
2) somebody was bound to do it sooner or later
3) it will be the basis of some very powerful post-production (and eventually live) audio processing tools

The logical conclusion of this path, though, is digital alchemy of sorts - except that it kind of works where classic alchemy never did. The end product, though, somehow lacks a certain authenticity that I find key to human artistic expression.

In the upcoming flame war, I'm declaring my early support for the Neanderthal viewpoint.

 

Hooray! I don't have to practice anymore!!!

 

Pitch Correction

It's been around a long time.
It happens all the time.

Put it in the trade publications?
No need.

Don't you guys watch the Simpson's???
It's the cheesiest way for someone with no talent to sound better.
IMO, it should be banned.

So what happens when you hear a pitch corrected cd and then the person shows up singing off key?
LOL.
The jokes on you for paying to see them but the bigger jokes on them for making a fool of themselves.

Bart knew. (not a bad show)

 

show me a modern album (that sells more than 50K copies) that contains ZERO pitch correction and I'll show you an engineer who can keep his mouth shut.

why ban it if it is cheesy? if it is cheesy, it will die naturally.

in the upcoming heated discussion I side with those who say "it is just another tool. Music will always require talent and time... and jeffscreamedcorn is so dang talented he don't need no practice no how.

Melodyne can be used for a variety of purposes including pitch correction, revoicing etc etc etc.

 

This doesn't look like a tool for those of using Garageband and Audacity. I'm sure there's a lot more going on than that, but yeah, it's a tool. If I want to use it, it'll probably take as long or longer to master than something like ProTools. I could probably spend the next few years in school studying that application exclusively. There's been a huge gap between how music sounds live and how it sounds in the studio for a very long time now (about as long as there have been studios, really) from tricks of space and instruments' physicality to tricks of tape to tricks of digital manipulation. I feel a little differently about running tools like this on a live sound rig, but that thread will come in its own time, I'm sure.

 

Neo,
I have been messing with it for about a month... the functions in melodyne will get put into things like GB eventually. it will be a simpler version, but it is only a matter of time.

you'll be surprised at how fast you are able to use melodyne if you have used logic or antares stuff... pretty straightforward and so deep you'd never know it all...

 

I see it going well beyond rudimentary applications like 'pitch correction'. I view it as an interesting approach to composition.

You could literally start with a single chord a triad even and evolve it through all kinds of permutations. Of particular interest to me is the notion that you could for instance compose and then 'play' a guitar piece that would actually be impossible to execute on a real guitar and have it sound completely real. You could literally create a symphonic orchestration totally within the voice of a guitar for instance and have it sound totally real. I find that fascinating.

Why not I say!!

 

symphonic orchestration with the sound of a guitar? sounds a bit familiar.. Brian May wrote his multiple-part guitar arrangements of God Save the Queen, and it sounded a bit like an organ due to the effects he was using. But I'm sure with this kind of sound manipulation, you could do some crazy stuff. Though there's the problem of playing it live... but I guess air guitar does the trick, eh?

 

Ding ding ding, we have a winner! My sentiments precisely.

While I have not yet utilized this technology, I've been intrigued with it since I saw the video a few months back, and given opportunity to do so, I certainly would embrace this technology. As I understand it, there are no digital artifacts involved in the process.

Here's a scenario: you're a hired gun for a session, and you're cutting on-the-fly. You play with good meter, feel, and dynamics, and you're given a "Hail Mary" rough mix after the date, OR, you hear a mix of additional instrumentation after-the-fact of cutting your bits. You say to yourself, well dang, in retrospect, I'd have voiced 'that chord' differently. Unfortunately, time and budget constraints are in effect, as per usual. Sound familiar? In this scenario, Why on Earth WOULDN'T you embrace this tool?. Man, I sure would, as long as the aforementioned "human" transients were unaffected.

One for the purists: Do you print your ambience and choose "your tone" with your track, or do you leave it to engineers to add their goop of choice and re-amp, after you've went your merry way? Given the latter, what's the difference between that and what we're talking about with this technology?

 

Is it me????

Almost every lousy YouTube instance just sits there with its little circle going round and round for me, forever — DOING NOTHING!

It's not just here at TDPRI. It's everywhere, it's everywhere! I'm so frustrated, I could just scream!

"Eek!"

That's better.

 

Reply to Kerry Vance

Kerry Vance

You're right. LOL. (They sure don't call me lazy for nothin')

Now that I saw what you're all talkin' about, I still think the same way.
It's no different than with the vocals.
Cheap tricks. Well, maybe expensive ones. Don't know.

IMO, it's a cheesy way around doin' it right.
In my book, that means doin' it real.

I'd really rather hear it live anyway.

Da** I need to get some sleep.

 

Well, so much for that argument...