Read more on BT's Emotional Technology in this online exclusive here
In dance music, there is no greater perfectionist than BT. The L.A.-based producer, DJ, remixer, film scorer, artist and pioneer of progressive trance and house is constantly taking a microscope to his sound, and the jack-turned-master of all musical trades can do just about anything in the studio. In fact, in talking with BT, unless you're at least at a master's level or beyond in music theory and technology, you can expect your eyes to glaze over on occasion. Even Remix encountered moments in the conversation when we failed to follow and fell into a Homer Simpson — like stupor. BT: “I did some granular synthesis and time- and formant-stretching where you stretch the formant independent to the fundamental, but I had to do a fast Fourier transform analysis of the part I was manipulating.” Remix: “Huh?”
BT, aka Brian Transeau, has had plenty of practice to arrive at this mad-scientist level of music devotion. BT's obsession started with playing piano at age 2, eventually leading him to study at the Berklee College of Music from '89 to '90. After dropping out and moving to Los Angeles to shop his demos, he returned to his hometown of Washington, D.C., without a deal and co-founded Deep Dish records with his high school friends Ali Shirazinia and Sharam Tayebi of Deep Dish. BT's tracks found their way to the ears of British DJ Sasha, and things took off from there. Since then, BT has worked as a film scorer (Under Suspicion, Zoolander, The Fast and the Furious and the upcoming Monster), a remixer (Tori Amos, Madonna, Sarah McLachlan, Seal), a producer ('NSync's “Pop,” Sting's “Never Coming Home,” Britney Spears' “I'm Not a Girl, Not Yet a Woman”) and a solo artist (Ima [Perfecto, 1995], ESCM [Perfecto, 1997], Movement in Still Life [Nettwerk, 2000]). He has also performed at clubs and festivals, such as Coachella and Glastonbury, around the world.
The former D.C. native, whose formative influences range from Stravinsky to Depeche Mode, recently added a new title to his resume — vocalist — with his fourth studio album, Emotional Technology (Nettwerk, 2003). He sings on six tracks, with additional guest vocals by actress Rose McGowan, Girls Against Boys' Scott McCloud and 'NSync's JC Chasez. Meanwhile, BT plays everything from bass, drums, guitar and synths to esoteric sound-design boxes such as Symbolic Sound's Kyma. He also hooked up with Guns N' Roses (the new GN'R, that is) members Brain (formerly of Primus, drums), Tommy Stinson (The Replacements, bass) and Richard Fortus (Love Spit Love; guitar, cello), as well as DJ Swamp (turntables), to record instrumentation.
In addition to knowing his way around the studio, BT is quite the professor of technology. He has hosted clinics at the 2003 Winter NAMM convention and the WMC Remix Hotel in Miami for Ableton Live and Propellerhead Reason. And he is currently writing a book about time correction due out next year. So without further ado, Remix will step aside while BT kicks some knowledge about songs on Emotional Technology. If you start to feel the words flying over your head, check out the sidebar “Huh? What? Some BT Terms Explained.”
“KNOWLEDGE OF SELF”
This is a collaboration I did with Guru of Gang Starr. I actually wrote three tracks for him to do vocals over, but he called me back and was like, “Dude, the tracks are too fast. We usually flow over something that's, like, 90 beats per minute.” All the tracks I sent him were over 130, so I took the track and slowed it down in Live and sent it back to him. When we got his vocals back, I did a very long and arduous vocal comp where every syllable was crossfaded up to the music, and it ended up being 132 bpm. The thing that's really cool with Live is, if you have a time-corrected piece of audio material, then warp marking is so easy because you basically have to indicate where the downbeat is. You can do every eight bars; you can do every downbeat. You just have to make sure that it's not drifting, but if you're using time-corrected material, you can literally put in two or three points, and it won't drift at all.
To give you an example of how complicated time correction is, I'm writing a book on it right now. It will be easily over 300 pages just because it involves different types of time compression and expansion; different styles of cutting both on- and off-axis; different styles of shelving, EQing, crossfading and compression of everything to get the right sound for one of 50 different types of sounds. For example, time-correcting piano cannot be done using cuts; you have to use either [Wave Mechanics] Speed or [Serato] Pitch n' Time, depending on the harmonic material at any given point in the performance. And you have to do it using exponential crossfades. It's not possible to do using cuts. For example, something that has a pre-attack, like a guitar, to leave the integrity of the pre-attack, it's extraordinarily sophisticated, and it involves the body of the sound and then grafting back on the pre-attack.
“Knowledge of Self” happens in three discrete sections. There's a beginning point of the song where the beats, keyboards and Guru's vocal get going, and it's this really cool, chunky, techy, house-break sort of vibe where the kick drum is on the one and the three, as opposed to the nu-school breakbeat sort of pattern where it's “one and a three e and four” [“one e and a two e and a three e and a four e and a” being the whole 16th-note count in a single bar]. The second part of the track introduces some really cool lines done with Kyma, Virus Indigo, [Big Tick] Rhino and the granular synth in Reason. And then the last part of the track goes into a more traditional breakbeat thing with a huge, nasty, distorted bass line and a kick that doesn't fall on the one and three.
The first part of it I wrote on tour last year in my tour bus on a six-string Yamaha bass using the mic input of a [Kurzweil] K2000 for distortion. It's something I highly recommend. Richard Fortus from Guns 'N Roses, who played some of the guitars on the track, was like, “Dude, that is the dopest distortion I've ever heard.” That track started as a verse and a chorus sort of riff. I decided I wanted to make it a tongue-in-cheek sort of thing, like Missing Persons or Oingo Boingo. So I wrote a lot of lead lines for that track on old synths like the ARP 2600, Micromoog, Polymoog and the [reFX] Beast VSTi software synth.
I was looking for a vocalist to do a sort of spoken-word rant about being an actress in Los Angeles, and I had some mutual friends in common with Rose McGowan. So I got together with her, and I was like, “I was thinking of a spoken-word thing.” But she was like, “No, I can sing.” So she did a vocal. And my friend Scott from Girls Against Boys did a vocal for that, too, and I sang the chorus. One of the interesting things about that track is taking and time-correcting loops in the song at different templates. So there are loops toward the end there that are swung fairly heavily. But the template for them was pushing the “e” [one e and a two e and a] of the beat 375 samples, pushing the “a” 550 samples. It's not a traditional swing where you just push the second and the fourth of every set of four 16th notes. It pushed the two out of every set of four 16th notes disproportionately to the four. It's a really interesting groove. And then that going against more traditional playing, like 220 or 230 samples, makes for a really cool feel on that track.
“DARK HEART DAWNING”
I wrote this song on an Alvarez acoustic baritone guitar that Richard Fortus had made for us. You can tune it down to an A or even a G, and it still sounds just fantastic. The first thing I did with it was program these dubby-style beats and played a Yamaha six-string bass on it. It's through-composed — it's definitely not a verse-chorus-verse-chorus sort of thing — and it's reminiscent of Talk Talk's “Spirit of Eden.” I mixed the song in sections because each section had its own sort of characteristic sound. And what I do with a lot of the songs from the album is, I would mix down to stem. It's a way I've found of working now, especially the way I do stutter edits; all of that kind of stuff is really easy to do when you're working with stems. Stems are something we use in film when you mix multiple channels to a single stereo audio file. So you take a track of 64 tracks and pare it down to eight tracks. So you have your drums, your bass, your vocals, your keyboards, your effects. … It makes for a much easier way of working. I had about five different [Emagic] Logic arrangements going and a couple of Pro Tools arrangements going, and all of them were submixed to 8-channel stems and then all pulled into one Arrange [window] in Logic. It's cool because, as opposed to blocking things out and doing global compression, you can do global compression stems, reline stuff up and re-time-correct it.
Mike DiMattia and Carlos Vasquez, who work in my studio, spent a month-and-a-half just time-correcting the guitar, bass and drums. We time-corrected every 64th note of 64 tracks of audio for 10 minutes. So it took six weeks, working every day. I went into the studio and cut this with Brain, Tommy Stinson and Richard Fortus. It's an incredible 10-minute performance, and I wanted to keep its integrity. But to keep the emotional integrity, I realized that you weren't going to do it by grabbing two-bar loops and then changing the turnaround on bar 8 every time. To make the loops work with it, everything in the song had to be time-corrected or else you've got a bunch of flamming, and it just sounds like ass.
“THE GREAT ESCAPE”
Caroline Lavelle and I were introduced by Peter Gabriel, who lives close to her. I love her voice — and she's an amazing cellist — so we got together and wrote a song at her house. The poem that Caroline wrote was about someone that she's had this unrequited love affair with for years, this painter who lived in her small town. One day, the police came to her town and tried to arrest him, and he fled. It turned out that he was one of the biggest heroin dealers in England. He came to America, and he was fleeing across Canada. He would write her these e-mails. Some of the lines in that song are just so beautiful. She says, “I see the sparks fly upwards and feel the warmth from all the of the burning of his bridges.” It's so beautiful. And some of those lines are things that we poached directly from his e-mails. It was like reading Henry Miller's letters to Anaïs Nin — the most beautiful, romantic letters I've ever read in my life.
So I recorded probably 120 different takes of Caroline singing. That vocal comp took three weeks. But with great singers, all of the takes are very similar, so you get into microfine detail of finding the best thing. My techniques for lining up vocals are as intricate as my time-correction techniques: very intense. I'll take a vowel sound, and I'll graft it on to a non-time-stretched plosive, like a p or a g, and then I'll put a consonant from another word at the end of it. Say you're trying to make the word can. You go and find the “aah” sound, and you make it as long as you need to with Pitch 'n Time or in Speed. I'll make it the pitch that I need to do a background to, say, make a perfect fourth. I'll find an “aah” that's maybe a minor second away from where I want it to be. I'll pitch-shift it up 100 cents and time-stretch it so that it's long enough. Then, I'll find a “cca” from another word. It could be in the middle of a word. I'll usually time-stretch the end of it on a linear slope, so the beginning of the word is not time-stretched; the end of it is time-stretched by 100 percent. So in other words, you're gradually trying to time-stretch the “cca” sound. I'll graft that into the “aah” sound, and it sounds much more human. But you have to vary the length of your crossfades and try linear-exponential or converse-exponential slope to get the “caaa” sound. And then I go find and “n.” I'll take the end and reverse it and then leave the beginning time-stretched and the end stretched by 100 percent using a linear curve, and then I'll flip it back so that it's time-stretched at the beginning of the sound, so it will morph nicely from the “aah” to the “n.” There were so many background parts that I wanted to use Caroline's voice for that she didn't sing. So I created so much of the background vocals for that song out of pieces. That's a technique that started with me when I was doing “Blue Skies” with Tori Amos. She never said the words blue or skies.
Also on “The Great Escape,” Richard Fortus played marcato cello lines with a pen, which sounds amazing. I wanted to do an aleatoric section in the middle of the song. There's no violin, viola or contrabass: The entire thing is cello. So Richard did 30 different takes. The cool thing about the aleatoric section is, there are some really cool formant time-independent stretches — granular vocal riffs over the aleatoric section. The other cool thing is the dub-style-effect-feeding-back-into-itself thing that a lot of the reggae guys did. I lined up all 30 tracks of cellos so that all of the open fifths at the end of this large crescendo that started at mezzo piano and ended at triple forte all lined up. After that, I did an effects send where I made a composite effects bus for all 30 tracks that I sent to the same effects bus, which was a reverb, the Echo Farm plug-in and some other effect to make it sound swirly. So I fed back that bus into itself and then automated the feedback. It sounds like this swirling, building cloud of distortion behind the aleatoric strings.
I was on the Eurorail from Paris to London, and I had Logic open and a bunch of different VSTis, and I was writing with my [Midiman] Oxygen 8. I was like, “Wow, I think this poem that I was working on would work really well with this part.” I started singing it in my head, and I was like, “Oh, my God, I need to record a vocal.” I went into the bathroom on the Eurorail, plugged the headphones into the mic input on my iBook — using the headphones as a mic — and sang the whole vocal into the laptop, parts of which actually made it to the final version of the song. And then, after London, I went down to Brighton, and I was working with my friend Jody [Wisternoff] from Way Out West. We started writing more music around what I'd written on the Eurorail. It's the first time that I've ever written a track while moving. I wanted thematically to keep those feelings of motion of the track, as it was written in motion. So I found some recordings of a train. So that track features sounds that I time-corrected down to 92nd notes. I call it “nano-correcting,” which is correcting unreal note values, so instead of just correcting up to 64th notes, I've corrected 128th notes, 512th notes and what have you. And the notes are so small that they end up sounding more like stutters. And you go, “Why is that train happening in time with the beats?” So that track has a time-corrected train with breathing from my BCD, my breathing apparatus when I scuba dive, and a thunder sound using a kettle drum. And it has this phenomenal granular edit done to the vocal with VSTis. A lot of those [software synths] were from places like www.sharewaremusicmachine.com.
“LAST MOMENT OF CLARITY”
I was in Florida with one of my really good friends, Bill Hamel, and we brought Mike who works in the studio. Mike doesn't get out of the studio a whole lot, so he went kind of nuts. In fact, we were all downtown at some bar, and we lost Mike. He went missing, and we didn't find him for, like, eight hours. So Mike had way too much to drink that night, wandered off and found himself walking on the freeway, and a cop pulls over and was like, “Dude, what are you doing?” And he said, “Man, I'm fucking partying it up!” And the cop said, “I should get you back somewhere. What are you doing here?” And Mike said, “I'm in town working with BT.” And the cop was like, “Dude, I love BT! Can I take you back to where he's working?” That's how we finally found Mike. The name of the song comes from a joke from Mike. He was like, “I remember my last moment of clarity that weekend.” And it just seemed to fit this beautiful stream-of-consciousness rant vocal that Karina Ware did. I played baritone acoustic guitar; some Peter Hook-y, New Order — type bass; and some straight-up Everything but the Girl — type house beats.
The cool thing about doing this song was, a couple of days after we found Mike, there was this beautiful flash storm in Orlando. I had Bill's MiniDisc player, and I went outside to record the storm. I was standing in his front yard with headphones on, and I could hear Bill listening to the song in the studio from outside. So you can hear the track off in the background in Bill's house, and I walk into the house. And then in Kyma, I did a freeze warp. So I did a fast Fourier transform analysis of the rain and walking into the house, and I did a fast Fourier transform analysis of the track. And there's a warp over 16 bars out of the rain and into the house. So it's like you're teleporting into the song. And then we spent a good six days nano-correcting the rain. The rain in the song is time-corrected. It sounds fucking mental.
[Read more about BT's Emotional Technology atwww.remixmag.com.]
HUH? WHAT? SOME BT TERMS EXPLAINED
aleatoric: Music that employs the element of chance in the choice of tones, rests, durations, rhythms, dynamics, et cetera.
amplitude: The highest or lowest point of a sound wave.
comp: To make a composite. To cut-and-paste together the best parts of multiple takes of one track.
crossfade: To combine signals so that one sound fades out as another fades in while maintaining a consistent volume.
envelope: The variation of maximum amplitude over time.
fast Fourier transform analysis: A quick form of a Fourier analysis, which is a mathematical technique to transform a mathematical function from one domain to another. Applied to audio signals, Fourier analysis transforms a waveform (a signal in the time domain) into a spectrum (a signal in the frequency domain) that reveals the frequency and amplitude of the component harmonics.
flamming: An audio occurrence in which one of the instruments used on a rhythm track strikes slightly behind the others.
granular synthesis: Overlapping and manipulating the placement (as well as frequencies, envelopes and wave shapes) of many slices or “grains” of sound — each ranging from 2 to 200 or so ms — to create complex and time-variant sounds.
formant: A resonant peak in a frequency spectrum.
fundamental: In a sound comprising many tones (including frequency, amplitude and phase), the fundamental is the lowest frequency in pitch.
marcato: Marked, stressed, emphasized, often with respect to the melody that is to be made prominent.
resonance: A natural vibrating frequency stimulated by an outside force of the same frequency. As one frequency approaches the other, oscillation occurs, which reaches a maximum amplitude at the natural resonant frequency.
through-composed: Creating different music for each verse.
warp marker: In Ableton Live software, a warp marker forces the software to arrive at a specific point in the sample at a specific time. Using warp markers, you can sync an audio sample with a chosen tempo and “warp” a sample's rhythmic flow to change its feel.
Lately, BT has been performing live remixing sets rather than just mixing from song to song. He explains how he does it:
“I'm using three titanium laptops, one of which I started running Logic on, and on the other ones, I run Live and Reason. I've got a bunch of warp markers and key tracks that I have in Live. And for my own tracks, I have a ton of separates, so I'll have a version of a track without a kick drum. As opposed to beat-mixing, say, Meat Katie with Dylan Rhymes, I'll take Meat Katie and then write a [Sonic Foundry] Acid line on the fly, and I'll beat-mix that in. I'll drop a breakbeat in Live, I'll write a bass line in Reason, and I'll beat-mix that with Meat Katie. Also, I'm taking a lot of old songs and mixing them with new songs, like mixing a breakbeat from ‘Trans-Europe Express’ by Kraftwerk with ‘Another Brick in the Wall, Pt. 3’ by Pink Floyd and something by Dylan Rhymes. So, sometimes, I'll have five or six tracks playing at once, shelved and filtered in various different ways. It's like juggling or plate spinning. Oftentimes, if I know I'm getting toward the end of the track at the beat-mix section, I'll put that fader in record, record a two-bar loop and then allow that loop to go for 10 minutes as I'm beat-mixing in and out other songs, other sections and other keyboard parts that I'm writing. You have to be so on-point to remember what you've got going. It gets so dense sometimes sonically that I'll go, ‘What's that hi-hat part from?’ I love doing it because it's very challenging.”
Read more on BT's Emotional Technology in this online exclusive here