Making a Song

I’ve learned to make music through reading, experimentation, and by asking questions of people more experienced than myself. One of the things I’ve learned along the way is that there is more than one way to go about many parts of the process. I’m not sure if my process is unique, rare, or the norm, but here is my process when it comes to making a song. Keep in mind that this is a general process. Sometimes the order of the steps is different, steps are skipped, or steps are added.

1. Initial Creation – A song has to start somewhere. Sometimes it will start with music, sometimes it will start with lyrics, sometimes lyrics and music will be created simultaneously. Creating the music involves choosing instrument sounds to use, and programming the notes into a midi sequencer.
2. Recording – Once the instrumental and vocal parts are written, I record the vocals. Usually, at this point, it’s just a rough version – especially if I’m the one doing the vocals. This rough take helps me hear how the music and the vocals mesh together, it gives me a chance to refine them so that they take even better advantage of each other, and it helps me learn the words for an improved take later on.
3. Rework the music – As I said, once I have the rough vocals, I can make the music and vocals work together even better. This is the point where I cut instruments out to highlight lines, add accents with drums, and rewrite parts of the lyrics and music if necessary. Because of this step, I never consider music I write complete until I have a rough version of the vocals and get a chance to rework the music around them.
4. Record Vocals – Normally when I make a new song I stop after step 3. Once the rough draft is down, I can come back anytime to finish the song. In the meantime I listen to the rough version a bunch of times and maybe even memorize the lyrics. Now, when it comes time to finish the song, my familiarity with the song (even if it’s not memorized) allows me to record a better vocal performance, more suitable for a final version.
5. Record the music – At this point, the instrumental is still midi – a set of instructions that, when combined with a hardware or software synthesizer, will replay the music. In order to add whatever processing is needed (EQ, compression, etc) I need to turn these midi tracks into traditional audio files. When I use software instruments, this is as easy as a few clicks of the mouse, but usually, my songs have a significant amount of instruments that come from my E-Mu Pk-6. When that is case, I have to record each instrument to a separate track. This step tends to be a pain in the ass. If I have a 4 minute song with 7 instruments played through the PK-6 (a typical scenario), it will take 28 minutes (4 minutes x 7 instruments) to record all the instruments. Add to this the time it takes to set up each track before I record it and the time it takes to redo tracks when there is an error (when dealing with midi, my computer frequently hiccups, throwing everything out of time) and the time to record the instruments grows – sometimes to more than twice that 28 minutes. Because I have to catch errors as they occur, I have to be nearby throughout the whole process. What makes this all so annoying is that this is a purely technical task – it involves no creativity and little thought. This is the type of busy work I would give to low level employees if I had the luxury of employing people other than myself.
6. Mix – Now that I have all the elements, I can mix them to make them all sound right together. Mixing involves EQuing and compressing the individual tracks, adding reverb and delay where they are needed, and applying any other effects and audio manipulation that is necessary. Sometimes I will mix with the rough vocal tracks and then carry out step 4 after this step.
7. Extras – This is actually a new step that I plan on incorporating into the process. Sometimes ambient noises or strange little background sounds can make a track come alive, but looking through a ton of sound effects while I’m trying to construct a track would be an interruption to my creative process. Instead, as one of my last steps, I plan to search for an add the subtle sounds that will help make the track more dynamic.
8. Fix – Throughout the process, I listen for parts of the song that need fixing or improvement, but at this point I try to be extra critical and analytical. If there is anything wrong, I go back and fix it.
9. Master – Now that everything is done, the track can be mastered. Ideally, I should be sending tracks off to a mastering house and paying them a heap of money so they can make sure my levels and overall EQ are well-suited for mainstream play, but for money reasons, I do this step myself. This is the final polish before I present a track to the public.