Out of many, one.
March 23, 2010 9:58 AM Subscribe
185 singers, 12 countries, one conductor -- all online. Grammy-nominated composer and conductor Eric Whitacre put out a call for singers on his blog in July of 2009. He then posted the conductor track for his piece "Lux Aurumque" and gave instructions, including how to audition for the brief soprano solo. Recordings trickled in on YouTube over the next few months until the January 1 deadline; the results were posted on March 22.
See this previous post on the blue for more examples of his work.
His opera Paradise Lost: Shadows and Wings will be performed in concert on June 15 at Carnegie Hall... and if you're a singer, you can participate.
See this previous post on the blue for more examples of his work.
His opera Paradise Lost: Shadows and Wings will be performed in concert on June 15 at Carnegie Hall... and if you're a singer, you can participate.
Freaky.
posted by ReeMonster at 10:12 AM on March 23, 2010
posted by ReeMonster at 10:12 AM on March 23, 2010
Wow, that's really impressive. It's a nice piece and well-assembled from around the world. I was particularly impressed by the lack of ambient noise from all those webcams, and the nice reverb (both were likely the result of post-processing, I guess). It seemed like Maestro Whitacre was a bit of a ham but I'm not really used to seeing conductors from any angle but behind; I guess the tight shot just seemed unusual to me.
posted by The Winsome Parker Lewis at 10:15 AM on March 23, 2010
posted by The Winsome Parker Lewis at 10:15 AM on March 23, 2010
That's exceptional.
Congrats on your first post, Madamina!
posted by zarq at 10:27 AM on March 23, 2010
Congrats on your first post, Madamina!
posted by zarq at 10:27 AM on March 23, 2010
Kind of like "Queen (A Portrait of Madonna)" (excerpt), but on a much larger scale. Cool.
posted by cog_nate at 10:30 AM on March 23, 2010
posted by cog_nate at 10:30 AM on March 23, 2010
The Winsome Parker Lewis: "the lack of ambient noise from all those webcams, and the nice reverb"
You can do some pretty extreme noise reduction with fft analysis/resynthesis. The side effect which makes it harder to use is the time smear (similar to the quantum mechanics speed/position problem, audio processing has a time/pitch problem - the more exact you get your frequency analysis, the worse your time data gets, and visa versa). Time smear is also known as reverb - getting rid of the background noise without reverb would have been much more of a technical feat.
posted by idiopath at 10:45 AM on March 23, 2010 [1 favorite]
You can do some pretty extreme noise reduction with fft analysis/resynthesis. The side effect which makes it harder to use is the time smear (similar to the quantum mechanics speed/position problem, audio processing has a time/pitch problem - the more exact you get your frequency analysis, the worse your time data gets, and visa versa). Time smear is also known as reverb - getting rid of the background noise without reverb would have been much more of a technical feat.
posted by idiopath at 10:45 AM on March 23, 2010 [1 favorite]
Who wants to walk me through what kind of tech/programs I would want/need to participate in something like this?
Then who wants to get together and sing stuff through the Internet?
posted by greekphilosophy at 10:55 AM on March 23, 2010
Then who wants to get together and sing stuff through the Internet?
posted by greekphilosophy at 10:55 AM on March 23, 2010
idiopath: Fascinating! I'm amazed at what can be done with modern audio processing. I used to do the basics with Adobe Audition back when I worked in radio (splicing, multitracking, time stretching, and some of the canned effects) but never dabbled with the really advanced stuff. I'd love to get into it someday, budget allowing.
posted by The Winsome Parker Lewis at 10:56 AM on March 23, 2010
posted by The Winsome Parker Lewis at 10:56 AM on March 23, 2010
I love Eric Whitacre. In fact, the reason I got into composing music was pretty much all owed to Cloudburst. (In fact, after having some "your my idol give composition tips to a young composer" correspondence, he sent me - as a Christmas gift - an autographed score to Cloudburst with a really sweet note.).
In any case, what a great composer with a great story (he was literally walking down the halls of UNLV - from which he took seven years to get his undergrad - when he walked past the band rehearsal and was like, "cool. what a good sound." He proceeded to go home that winter break and compose Ghost Train.) Cool concept.
I've been so grar-y that I haven't been in a location that would allow me to see Paradise Lost yet. Wish I could be in NYC in June. Sigh.
Great post, thanks!
posted by Lutoslawski at 10:59 AM on March 23, 2010
In any case, what a great composer with a great story (he was literally walking down the halls of UNLV - from which he took seven years to get his undergrad - when he walked past the band rehearsal and was like, "cool. what a good sound." He proceeded to go home that winter break and compose Ghost Train.) Cool concept.
I've been so grar-y that I haven't been in a location that would allow me to see Paradise Lost yet. Wish I could be in NYC in June. Sigh.
Great post, thanks!
posted by Lutoslawski at 10:59 AM on March 23, 2010
*you're still having coffee
posted by Lutoslawski at 10:59 AM on March 23, 2010
posted by Lutoslawski at 10:59 AM on March 23, 2010
greekphilosophy: "Who wants to walk me through what kind of tech/programs I would want/need to participate in something like this?"
Some guesses: for laying out the source material on a timeline: protools or cakewalk. For the audio processing an FFT based noise reduction VST/AU plugin, a lowpass resonant filter plugin, likely an autotune plugin as well, and a reverb plugin. There is crossover between filtration / noise reduction / reverb / pitch correction, so it could be patched together in a number of different ways, and there are single plugins which do more than one of those operations all in one go. I am fairly certain that a) there was a seperate fft based noise reduction as the first effect on each voice (the artifacts from that are hard to mistake) and that b) there was a discreet reverb plugin, and that was the last effect in the chain, and every voice was fed into it, but that is pretty much standard so not much of an insight. You would want to experiment a bit with filters / pitch correction on the voices in between to duplicate the sound he got.
Sadly I can't go into much more detail than that, I design my own effects and typically don't use plugins, and the ones I do use wouldn't be your best choices on Mac or Windows anyway.
The thing with how powerful audio processing software is, is that he could have started with a bunch of yowling felines, and with a bit of care gotten something as pleasant as the current recording. In many ways. if you care about skill and craftspersonship, the people who developed the effects algorithms and the UI for the software have probably done more to earn our respect than anyone else involved in this project. I don't mean to disrespect the singers and the composer, by all means use the best tools available for your craft, but rather I think that a significant amount of the work that goes into something like this was the work of the toolmakers (ideally Chowning, Fourier, Strong, Xenakis, and Gabor would have the sort of name that Stradivarius does).
posted by idiopath at 11:46 AM on March 23, 2010 [1 favorite]
Some guesses: for laying out the source material on a timeline: protools or cakewalk. For the audio processing an FFT based noise reduction VST/AU plugin, a lowpass resonant filter plugin, likely an autotune plugin as well, and a reverb plugin. There is crossover between filtration / noise reduction / reverb / pitch correction, so it could be patched together in a number of different ways, and there are single plugins which do more than one of those operations all in one go. I am fairly certain that a) there was a seperate fft based noise reduction as the first effect on each voice (the artifacts from that are hard to mistake) and that b) there was a discreet reverb plugin, and that was the last effect in the chain, and every voice was fed into it, but that is pretty much standard so not much of an insight. You would want to experiment a bit with filters / pitch correction on the voices in between to duplicate the sound he got.
Sadly I can't go into much more detail than that, I design my own effects and typically don't use plugins, and the ones I do use wouldn't be your best choices on Mac or Windows anyway.
The thing with how powerful audio processing software is, is that he could have started with a bunch of yowling felines, and with a bit of care gotten something as pleasant as the current recording. In many ways. if you care about skill and craftspersonship, the people who developed the effects algorithms and the UI for the software have probably done more to earn our respect than anyone else involved in this project. I don't mean to disrespect the singers and the composer, by all means use the best tools available for your craft, but rather I think that a significant amount of the work that goes into something like this was the work of the toolmakers (ideally Chowning, Fourier, Strong, Xenakis, and Gabor would have the sort of name that Stradivarius does).
posted by idiopath at 11:46 AM on March 23, 2010 [1 favorite]
Xenakis as in Iannis Xenakis? What's he responsible for as a toolmaker? I think of him as one of the first great artists to pick up those tools, for sure, but I didn't realize he'd done anything important for the technology itself (...which is, I realize, just an example of the problem you're complaining about).
posted by nebulawindphone at 11:58 AM on March 23, 2010
posted by nebulawindphone at 11:58 AM on March 23, 2010
I clicked on this expecting something neat. Quite unexpectedly, though, I wound up bursting into tears with the awesomeness of it. I've done voice recording work, and I know the truth of what idiopath speaks, but I apparently have a soft spot for people making one tiny contribution, lots of people, all over the world, that gets spun together into an amazing whole.
posted by KathrynT at 12:17 PM on March 23, 2010
posted by KathrynT at 12:17 PM on March 23, 2010
TWPL, I sing in a professional-level symphonic choir, and I gotta tell you, that guy is NOTHING in terms of hamminess. ;-)
posted by KathrynT at 12:21 PM on March 23, 2010
posted by KathrynT at 12:21 PM on March 23, 2010
I agree KathrynT - if you want hammy, I'll show you hammy. I was in SHOW CHOIR. *Jazz hands!* And then I'll show you a group of white kids singing really painfully precise gospel music. Meep.
posted by greekphilosophy at 12:32 PM on March 23, 2010 [1 favorite]
posted by greekphilosophy at 12:32 PM on March 23, 2010 [1 favorite]
nebulawindphone: "Xenakis as in Iannis Xenakis? What's he responsible for as a toolmaker?"
UPIC. the first really awesome GUI for computer music, and granular synthesis (well Gabor and Xenakis both claim to have invented granular synthesis, and UPIC was his idea but he was not the lead programmer, but I think he earns some tool maker cred for how early and how revolutionary those two things were).
posted by idiopath at 1:24 PM on March 23, 2010
UPIC. the first really awesome GUI for computer music, and granular synthesis (well Gabor and Xenakis both claim to have invented granular synthesis, and UPIC was his idea but he was not the lead programmer, but I think he earns some tool maker cred for how early and how revolutionary those two things were).
posted by idiopath at 1:24 PM on March 23, 2010
With his legacy in tool inventing and also composition using those tools, I guess Xenakis could be compared to Sousa (I hope somebody else finds that idea as hilarious as I do).
posted by idiopath at 1:27 PM on March 23, 2010
posted by idiopath at 1:27 PM on March 23, 2010
(these three comments really should have been just one comment, I just got back from a longish bike ride and I am a bit scattered)
The early composers of electronic music were by necessity also the inventors, or at least had some part in the invention, of their tools (which were often in some large part repurposing of tools meant for other uses): Stockhausen, Brün, Rissett, Cage, Hiller, Tenney, Tudor are a few off the top of my head. And then of course in the rock realm there there are Hendrix and of course Les Paul, who was smart enough to get a leading brand named after him.
posted by idiopath at 1:35 PM on March 23, 2010 [1 favorite]
The early composers of electronic music were by necessity also the inventors, or at least had some part in the invention, of their tools (which were often in some large part repurposing of tools meant for other uses): Stockhausen, Brün, Rissett, Cage, Hiller, Tenney, Tudor are a few off the top of my head. And then of course in the rock realm there there are Hendrix and of course Les Paul, who was smart enough to get a leading brand named after him.
posted by idiopath at 1:35 PM on March 23, 2010 [1 favorite]
. I am fairly certain that a) there was a seperate fft based noise reduction as the first effect on each voice (the artifacts from that are hard to mistake)
Can you describe them? Are they those background high-pitched fast staccato beeps?
posted by weston at 4:54 PM on March 23, 2010
Can you describe them? Are they those background high-pitched fast staccato beeps?
posted by weston at 4:54 PM on March 23, 2010
I had actually kind of forgotten that people sing in big choirs. Lovely and very "human".
posted by bonobothegreat at 5:02 PM on March 23, 2010
posted by bonobothegreat at 5:02 PM on March 23, 2010
weston: "Are they those background high-pitched fast staccato beeps?"
yeah, that is what FFT anal+resyn gives for sounds that it does not have enough partials to recreate - you get a similar thing with mp3 (especially on cymbal crashes), which also uses FFT as part of its data compression. You can do this on purpose and make white noise transition from sounding like a waterfall to a series of "computery bloops" as you reduce the number of partials more and more.
posted by idiopath at 5:19 PM on March 23, 2010
yeah, that is what FFT anal+resyn gives for sounds that it does not have enough partials to recreate - you get a similar thing with mp3 (especially on cymbal crashes), which also uses FFT as part of its data compression. You can do this on purpose and make white noise transition from sounding like a waterfall to a series of "computery bloops" as you reduce the number of partials more and more.
posted by idiopath at 5:19 PM on March 23, 2010
I came across the project on Whitacre's website right after he posted it last summer and thought what a fun thing it would be to try to participate--and then promptly forgot about it. I'm glad to have caught the results! Beautiful piece, and a nice concept, nicely done.
posted by ancientgower at 6:50 PM on March 23, 2010
posted by ancientgower at 6:50 PM on March 23, 2010
« Older Global Oil Reserves 'Exaggerated by a Third' | Oh The Inhumanity Newer »
This thread has been archived and is closed to new comments
I made a post on the blue! Go me!
posted by Madamina at 9:59 AM on March 23, 2010