Today on my blog, a “Making Of” special.
It all started with a conversation with a film student friend in Florida, who was making a documentary as a student project. He liked a song by Kenny Ascher and Paul Williams, called “The Rainbow Connection”that Kermit the Frog sings in “The Muppet Movie”. He wanted to cut a montage of images around it, but unfortunately, he felt that Kermit’s version would be a bit too cartoonish, and Kermit’s voice might distract. I told him I might be able to do something quick and dirty, pro bono.
I’ve been a Macintosh owner for a couple of years now … I still use PC’s for most of my daily work, but so many of the clients and studios with which I work are Mac-based, I really needed to join the fraternity. One of the most fascinating Mac applications, to me, is GarageBand. While falling well short of the versatility and performance of a professional system like Pro Tools, it’s quite fun to play with and has a lot of nice features. I’d been looking for an excuse to play with it, and this seemed like a great chance.
My first step was finding a starting point. I don’t play any instrument well, so I needed a basic performance to build around. This came in the form of a simple, general MIDI file found on a public “midi archive” web site. The site did not credit a performer or sequencer, so I don’t know who my benefactor might have been.
GarageBand is not a MIDI sequencer, so the performance needed to be converted into something with which I could work. It took a bit of searching, but I finally came upon an application called Dent du MIDI by a nice fellow named Bery Rinaldo. The program splits the MIDI file into tracks, one for each MIDI channel, and creates individual files that can be imported as GarageBand tracks. It worked flawlessly!
Next, I imported each of the tracks into GarageBand. The tracks come in as raw note data, so I then had to audition each track, determine what it was, and assign the correct instrument from GarageBand’s collection of synthesizers. In some cases it was necessary to transpose a track up or down an octave to improve the blend. Once this was done, I had a basic performance.
The arrangement was thin, so I added a few overdubs of my own using a USB MIDI keyboard and GarageBand’s sequencing feature. A string part here, a woodwind there, a better bass line, and some drums, and the song took on more of its original depth. I must admit, the editing features in GarageBand made this simple, and I’m happy about that because I am an exceptionally poor player.
Now, I had lots of instruments all making the correct sounds, and it was time to set up a rough mix. This was hard, even though it’s the part where I supposedly have the most skill and experience! GarageBand doesn’t have a mixing “console”, per se, but instead it has a feature where you connect dots on the track display with lines that represent changes in volume for a particular track. This took a lot of getting used to. Equalization, dynamics, and time-domain effects are also available, but similarly difficult to access and manipulate. What would have been a simple, two or three pass mix on even a simple console took about an hour of tweaking with this user interface. Of course, it might have gone faster if this hadn’t been my first time using it.
There, so the mix was reasonably together, and the only thing missing was a vocal. I’d decided to give it a try myself, so all I needed was a microphone. I have several very nice studio microphones, but the idea of setting up a Neumann in my living room seemed a bit absurd, and not at all in keeping with the amateurish nature of the project to date. Besides, I had no mic preamps in the house. Rummaging in my closet, I found an AKG headset mic of the electret condenser variety, and after plugging it into my little iMic mic interface and singing a few lines into it, I decided it would do.
GarageBand allows you to do an impressive amount of vocal processing. There are “presets” that contain pre-selected settings for EQ, compression, and effects, or you can start fresh. Being a purist, I started with a clean slate and applied a small amount of compression, some EQ mainly for presence and to overcome some shortcomings of the microphone, a little de-essing, and enough light reverb to smooth out my tone. I plugged my iPod headphones into the front panel of the Mac as monitors, and recorded three good takes of the whole song, not trying to sound like Kermit, but at the same time not trying to sound too confident either, going for the sort of tentative, understated feel that my friend said he wanted. In the process, I discovered the key of the original performance put a note or two out of my reach. GarageBand allowed me to transpose the whole piece down a half step with hardly any effort at all.
I listened to all three takes, and decided I needed to cheat. So, I chopped the takes to bits, created an empty fourth track, and assembled the best bits from the other three takes. After some quick editing to remove extraneous breath sounds and the like, I had a finished track better than I could have ever sung down live. Oh, and it’s not really cheating … it’s called “comping” and we do it in the studio all the time, you didn’t really think Paula Abdul could sing, did you? 🙂
At this point, I had managed to completely amaze myself at just how much I could do with this little program. All that was left to me now was the final mix and mastering. The mix was a breeze. There’s even a spot, in the last verse of the song, where the lyrics are “Have you been half asleep, and have you heard voices?” In Kermit’s version, the audio goes all “dreamy” (heavy reverb) as he sings those lines, and I liked that, so I cut out the two lines that needed the effect, dragged them to another track, and added a nice dreamy echo/reverb mix. The result was kind of cool, so I kept it.
Mastering was as easy as every other step had been. I just set the parameters for the master track — a little compression, a bit of stereo enhancement, and a dash of overall EQ — and clicked “Export to iTunes”. A few seconds later I was listening to the final product, to which you can now listen, if you’re brave:Â Â Scott’s Rainbow Connection
Next week: Scott builds a nuclear weapon using only materials found in any kitchen.
Permalink
Truly amazing, Scotty! And you have a great voice. If your friend in Florida doesn’t like it, then you’re not an honorary Englishman!
Permalink
And I’m a Dutchman!
Permalink
Superb. That was excellent. It really cheered up my afternoon.
Permalink
I didn’t understand teribly much of the technical stuff but the end result is superb!! Way to go Scott!
Permalink
Totally fab. I have a ridiculously large grin on my face after listening to that – suddenly I am six all over again. Wonderful.
Permalink
Oh, and you really must do more.
Permalink
I love that…one of my favorite songs and you sing it just right…I have been listening to it over and over. Thanks!!! You have such a fine voice….more, more!!!
Permalink
1. Come and fix my PC so I have sound
2. You seem to have an inordinate amount of spare kit lying around, for someone who has recently moved
3. You sound happy
4. You sound too clever by half
Permalink
Ooh, I can’t wait for the instructions on how to build a nuclear weapon. Perhaps the week after I could do a bit about how to make your very own clone.
(bacteria only, I’m afraid :))
Permalink
Wowzah! That dinnarf make me grin like a kiddywink, I tell ya!
Now, about this nookleear weapon thingy…
Permalink
Great job Scott! You hit the nail on the mark – after 30 seconds of playing, my officemate said “Um, why are you listening to Kermit the Frog?” Well done.
Permalink
That was lovely, had me crying and smiling at the same time!!
This song and the photo of you as a pirate really can not be reconciled!!
Permalink
Ooh bravo! Encore!
Permalink
Great job! Out of curiosity,anyone know the meaning of the song or wish to take a stab at it?
Permalink
The rainbow connection is the path that lies between us and our dreams.
Or its the gay bar down the street from me.
Permalink
Love it, you have a great voice. Thanx for sharing.
Permalink
I only stumbled across this by accident via a Google search for something else, but I have to say that I found the whole thing delightful — both the description of the process and the end result. Bravo! Encore! 🙂
— Jondar
Permalink
Great treatment of a classic. Good job!
Permalink
I agree with dan that the message of the song is that the rainbow connection is that path between us and our dreams. I might add that the other half of the message is that just wishing is not going to make dreams come true, despite the fact that’s exactly what most people are doing about their dreams. That is, if they listen to their inner voice at all to realize that they have dreams.
There are songs and stories about how people achieved their dreams, but then after hearing all that, our take-away is usually that they were born great, they had great breaks etc, and it’s not going to happen to us. Most of us just go about doing what we doing everyday and not do anything about our dreams.
Life is about taking deliberate and precise actions. Leaving dreams to fate, to prayers, to chance, etc, may be a romantic notion but not a practical measure.
My two cents.
Permalink
I have always like this song, you did a great job Scott. I think the line, Have you been half asleep and have you heard voices….is about out of body experiences ( OBE’S ), the first few seconds when you fall asleep.
Permalink
Outstanding!
I think I’m in luuuve!
Permalink
Scott, I’m both impressed and thrilled to have found your version of “Rainbow Connection.” I adore the song, and I wanted a Kermit-type (but not quite Kermit) rendition of it to play behind a photo slide show I’m making as a personal gift for my grandson’s 3rd birthday. Nothing commercial, but I still have my standards! My Internet search landed me on your Web site, and your recording is absolutely perfect. Thank you so much for sharing it.
Janie
Permalink
great job!