Wednesday, March 30, 2011

How to make Rebecca Black sound halfway listenable

Software program PaulStretch lets you slow down songs, transforming a hideously banal tune into something that sounds like a collaboration between Kate Bush and Mogwai

These days, two things seem to happen when rotten pieces of music achieve a certain level of notoriety. (Yes, I am about to mention Rebecca Black again, but bear with me.) First there's the outpouring of untrammelled fury over the fact that someone has dared to create something that seems to be culturally worthless. Then someone takes the track and timestretches it.

The notion of sitting at home making Rebecca Black's Friday last five times longer would seem, on the face of it, to be an act of self-flagellation. Uploading it for others to listen, surely an act of barbarism. But, as you'll hear, it's a beautiful thing. A shimmering, slow-shifting soundscape that transforms the hideously banal lyrics into a series of elongated vowels and hissing consonants. The music itself ? which is, at normal speed, a GSCE-type exercise in how two notes in a scale can fit reasonably well with a repeated sequence of four chords ? becomes complex, even fascinating. Passing a magnifying glass over something stultifyingly predictable can reveal minuscule, beautiful shifts in harmony and timbre that its creators would, it's safe to say, never have come up with in a month of Fridays.

This isn't new, of course. Back in August someone realised that Justin Bieber's U Smile actually contained some hidden beauty ? you just had to play it 800% slower. The software that's used to produce all these pieces is called PaulStretch, created by programmer Paul Nasca. In fact, it's not a strict timestretch; it's enhanced by a clever process of randomisation, where each tiny chunk of the music ? a hundred milliseconds or so ? is "smeared", digitally rebuilt and then placed back in sequence. While tens of thousands of people have sat back and enjoyed these ambient symphonies, from the well-known (Jurassic Park) to the more recently uploaded (Emmerdale) the credit really has to go to Nasca for producing a piece of software that effortlessly emulates ambient work that artists on the 4AD label 25 years ago would have spent weeks crafting. All we have to do now is, literally, click a button.

That took me two minutes; a Mike Sammes paint commercial transformed into something akin to Le Myst�re Des Voix Bulgares. It sounds lovely, but ... I don't feel as if I've created something. Because it required no effort.

There'll always be debate surrounding whether merely digitally processing other people's work is artistically sound. A lot of attention was given in the autumn to Daniel Lopatin, aka Oneohtrix Point Never, when he produced a loop of two bars of Chris de Burgh's Lady in Red ? probably the two least offensive bars of the whole song ? and was hailed as something approaching a genius. Surely that's easy, right? Well, I've just spent five minutes sticking a drum loop behind a section of Starship's We Built This City to show just how simple the process can be.

Unfortunately it's a bit rubbish. There's clearly a skill in making bad music sound good ? but at the moment I'd say Paul Nasca deserves the most praise. And I've not even heard any music he's made. Bizarre.


guardian.co.uk © Guardian News & Media Limited 2011 | Use of this content is subject to our Terms & Conditions | More Feeds


Jennifer Morrison Adrianne Palicki Amanda Righetti

No comments:

Post a Comment