Skip to main content

I asked a friend to describe Melodyne a new stand-alone audio manipulation software... He has a demo at home and saw it being demontrated at the Musik Messe in Frankfurt last weekend... Chris is from Belgium and has perfect English (of course!)

"yea right . . If only I could do it in flemish .
I'll give a more detailed report once I get my hands on my full operative version. for now I only am playing around with the demo. Which is fully functionnal but you cannot export / save your edits ... which IMHO is a very important factor because I want to know what it "really" does to your audio files once you export / render (?) them.
What I've come up with so far :
I important a vocal (44.1/16 - the full version can operate at 192/24) from a girl which I have been working with at the studio ... the kind that has everything / including a record deal but unfortunately needs autotune and some timing corrections. so ... after importing the 4 minute lead file and 6 backing vocal tracks (7 tracks in total and a mono reference music mix I started working on them.
first ... pitch / rythm detection. (I did this at home on my G4/733 ... using the soundmanager as a playback engine (the new 1.2 version is supposed to be PT compatible)
so ... detection ... this is where the computer takes a while ... 7 4 minute tracks ... took him about 20 minutes to detect rythm and pitch ... I told the computer the time and the bpm and it imediatly put the whole melody line in view without one single error. then I entered the scale the song was in.
here comes the autotune fun. . it puts the visual waveform relative to the scale pitch. you can immediately see what notes are too high or low. no need to even listen ... lol . then I selected "snap to scale" and voila ... all notes are pitch corrected when you click on them. and I know I have quite good pitch hearing possibilities. I can asure you ... it did what it had to do ... correct the pitch. 7 tracks pitch corrected in about an hour (importing and detection included so about 40 minutes of editing). even with autotune 3 in gra^hical mode it would have taken me at least 5 hours to get the same result. I even went further then I would have with autotune ... sometimes like with autotune ... you just hear it correcting the pitch ... that's where the formant editing comes in ... for a note pitched up ... dragg the formant bar a little bit down and it sound natural again ... just for fun I tried this pitching the whole thing up one octave ... and dragging the formants down about a fifth ... you hear it but damn ... it is good ...
next up the timestretching / timing correction. quite simple too ... scary simple even. you enter a grid mode and you just drag notes that are to short or too long to its correct possition. since it is multitrack ... you can see for the backings which ones start too early or too late ... and you just align them note per note.
It is supposed to be able to detect the "feel" of a certain track which one can copy and paste on to ather tracks / instruments ... even if there is not the same number of notes ... it is supposed to copy the "groove" on to the other pattern / track. since I only have the demo (1.0) and not a manual I didn't find out yet how to do that ... maybe it is even a 1.1 added feature ... but I've seen it work in Franfurt and it was quite amazing.
anyhow ... all the editing I did took me merely 3 hours to complete for a complete song. and that without the manual and spending like maybe 35% of that time trying to figure out how it works.
The hard part is that unfortunately I couldn't save my work ... quit program and poof ... everything gone . ... but I reasured myself that what I saw was not only an impressive demo but it realy works in real life situations too .
all for now ... more to come when I get to know it more and better and especially ... faster."
-
Chris Lambrechts
http://www.sos-recording.com

Comments

anonymous Wed, 03/20/2002 - 12:54

Originally posted by Jay Kahrs:
Geez, and I thought you had good news. I thought music had changed and we'd never have to think about using Auto Tune or such evil again because people could actually sing and play. Sigh... I guess it was only a dream.

Then his post would have read "Sell your DAW's and go back to tape!"

Guest Wed, 03/20/2002 - 18:00

Jay's post raises some very thought-provoking issues.

As an engineer in my own studio, I usually feel duty-bound (as well as ego-driven) to do whatever is in my power to make the best possible sounding recording out of whatever stuff comes my way. If that means auto-tuning a vocal, sample-replacing a kick, nudging a wayward snare hit, or whatever else, I rarely will hesitate - but only with the permission of the client of course. I wouldn't do it behind the client's back any more than I would sneak in a top pro without permission to rerecord a lousy bass track.

But the implication is still that it is cheating to make a product that sounds better than what the musicians could actually create in real life. I can see the point - there is a certain dishonesty in making a mediocre performer sound so good that the public assumes that they can sound that way in concert, for instance. It seems not far removed from Milli Vanilli.

But on the other hand, any time we run a track through a Pultec or a Fairchild or a Massive Passive or an LA2A or (name your favorite piece) aren't we still creating an illusion, albeit perhaps a more socially acceptable one? Is using a digital room simulator that much more "honest" than using Autotune? For that matter, how about punching in? Overdubs? Playing a tape backwards? Since when is comping a vocal being true to the performance?

I'm playing devil's advocate here. I don't actually think they are exactly the same, at least in degree. But we are all in the business of altering sound in some respects. While I hate to see high levels of production creating stars out of non-talent, I am in no position to accept artists into my studio only if they can pass an audition. I'd love to work only with top pros who always sing/play in tune and in time, or else have such great energy that it doesn't matter. But those of us towards the bottom of the food chain have to take what we can get. If I can make someone sound better than they ever did in real life, I'll be glad to do it.

Meanwhile, sign me up for that fuggin' Melodyne!!!

anonymous Wed, 03/20/2002 - 18:46

A recording artist and a live performer ,IMO, are two seperate things, although, most of the time they go hand in hand. A recording artist is some one that does the best they can to create the sound that is in their mind (or vision, or what ever). The recording artists uses whatever tools that are available to acomplish this goal. In todays times there are no rules. You do what ever you have to, to get the sound you are looking for. You use what ever tools are available ie. producers, studio musicians, gear, Autotune, samples loops. What ever it takes to express the the sound, or what ever is being communicated.

The live performer is a totaly different category, and IMO shouldn't really be compared. One is to entertain and the other is to have a tangible representation of the "recorded creation".

Just like a painter that is painting the trim on a new house. Is it cheating to use caulk and cover up the cracks that the finisher (the trim guy) left. No. You do what ever you can to get the best outcome.

Just a thought, of course, just my opinion.

... I think it would be good to have spell check on these boards too. If we did I would look a whole lot smarter.... LOL! .... or maybe it is just artistic interpretation, I really didn't mispell the words.... I .... ugh.... just said them different.... yeah! thats it!! ... to express my self...

Darren

MadMoose Thu, 03/21/2002 - 20:00

Originally posted by Julian Standen:
Gear gear gear...

But without the music there is no gear. Why can't we talk about the music for once? At the end of the day that's all we hear. Now, though things are flipping the other way and we see technology and gear running the sessions. While either one works, it makes my job more enjoyable when someone can actually play and I don't have to break out the big power tools.

While we can get session players or use Auto Tune to help someone communicate sometimes we need to ask what it is that's trying to be communicated. Why help them out? If someone is communicating their vision of music like a 2 year old rather then someone with a high school or college education why do we feel obligated to make them sound more mature? Let them sound like a 2 year old.

Damn, now I'm starting to feel all philosophical like Mixerman. Help!

anonymous Sun, 03/24/2002 - 17:51

When you punch in a vocal track you're trying to capture a better performance from the artist. When it's necessary to autotune a vocal track you're artificially creating a better performance than the artist is capable of producing. BIG difference.

Using autotune on background or harmony vocals or for a slightly out of tune instrument can be a great time saver. Using it on a lead vocal is fakery. As a vocalist I won't use it. If I can't sing it I do the best I can and live with it.
As an engineer if my client wants to use it ...well they pay my bills.

anonymous Mon, 03/25/2002 - 04:50

Originally posted by Cornflakeofdeath:
[QB]
Using autotune on background or harmony vocals or for a slightly out of tune instrument can be a great time saver. Using it on a lead vocal is fakery. As a vocalist I won't use it. If I can't sing it I do the best I can and live with it.
QB]

What about if you sang a great vocal take that was ruined by just one note being slightly sharp or flat? Of course, you can combine several takes to solve this simple problem but why not just use autotune? It saves you a lot of time and effort.

Also, aren't you forgetting the creative aspect of autotune? Saying flat-out you don't want to use it is a little short-sighted in my opinion. It can be used for more than just fixing shitty singing. Cher anyone?

Produceher Mon, 03/25/2002 - 16:03

One thing that autotune does do is stop those 8 hour vocal sessions where your just fixing pitch.

As a producer / engineer it gets difficult to tell if the performance is good if after awhile all your worrying about is pitch.

If I find a singer who just won't hold pitch, I'll just go for emotion. Make em sing 8-16 passes with lots of emotion and I can do the rest without losing focus.

It's kind of like those guys who EQ guitars for about 3hours before hitting record. That drives me nuts. 5 minutes at the most

anonymous Tue, 03/26/2002 - 02:11

Originally posted by phat-ass:
I finally just yesterday noticed the effect autotune actually has on the timbre of a persons voice. Now I know why Vocoders got so popular again. Or am I just making this up?

No you're right, I have such a love/hate relationship with autotune, I love the way it sounds, but I hate they way it's used on EVERYTHING these days.

ARGH!!!

Guest Tue, 03/26/2002 - 05:44

Originally posted by price wayne:
hi everyone,
i was wondering how the melodyne compares to the antares auto-tune pitch corrector.

sincerely,
wayne

Don't you love it when you're playing solo piano in some lounge, and inevitably some guy comes up right after you finished playing "Embraceable You" and says: "Can you play Embraceable You?"

Nothing personal, Wayne - just musing out loud. But it seems to me Chris did a really awesome job of answering a lot of your question in the very first post on this thread. If there are more specific questions you have that weren't covered, maybe you could ask them more specifically?

anonymous Tue, 03/26/2002 - 06:49

well, i just wanted to know,if you had a choice between the two,which one would you pick and why?also does melodyne make a plug-in pitch corrector like the atr-1 by antares?if that was mentioned before too,i am sorry.i am not too caught up on the specific terms so it might have slipped over my head.

sincerely,
wayne

Ted Nightshade Tue, 03/26/2002 - 08:07

I'm working with a very very talented composer, performer, keyboardist, vibraphonist, and singer (all ine one) who until recently has had real pitch problems, combined with vocal projection problems.
Well, we got to this point where it was, shall we invest in autotune or do you want to do it the best way, like you do everything else?
We were fortunate enough to find an exceptional vocal teacher, and within weeks, no more pitch problems, the mic pre is using 20 db less gain, and the whole performance is vastly improved.
The hangups that were messing with pitch were also messing with volume, dynamics, conviction, tone, expression, and every other aspect of the vocal performance.
I know not everyone is in a position to really address the root issues, but autotune and it's ilk are such bandaid solutions.
Not only were we able to get some great singing, but one artist's pysche is really getting some healing- which is leading inevitably to big improvements in all areas of her work.
Heavy stuff.
Ted

anonymous Tue, 03/26/2002 - 09:22

Originally posted by Renie:
I heard that Melodyne is going to be available to work with PT. anyone know about this?
I find it hard to imagine PT allowing another sequencer to be direct connected. It would be great if Celemony and Digi got together on this though..

version 1.2 is supposed / rumoured to add direct connect .... heard this both from a digi guy and a melodyne guy

anonymous Wed, 03/27/2002 - 12:40

Originally posted by Renie:
Chris

thanks for the news..do you plan to use the two app's together?

Renie

PT and melodyne ??? euh .... allready use ProTools and just added melodyne ... so ... yes I guess ....

Will have a nice story to tell in a couple of days ... doing some experimenting with melodyne and it's working FANTASTIC. I'll report it here when I'm done with it.

anonymous Wed, 03/27/2002 - 14:40

Sorry I meant to say do you plan to use the direct connect version when it comes out -so that the two app's are synced and integrated? -I was wondering how that might work -AFAIK Digi don't allow any other sequencers in da house yet so it should be interesting.

Look forward to hearing your thoughts on the programme.

Renie