Pages

Sunday, July 8, 2012

My Recording Process

I recently received some complimentary feedback on my latest album, Cherchez La Femme, from someone who has inquired about my recording process, so I decided to write a blog post about it. The nice note:
Damon – you asked, and now you shall receive. Your kind comments and question inspired me to blog about how I record, so maybe this will inspire you further. Welcome world to my explanation of my simple and efficient recording process – the one I’ve used for the 65 songs and 5 albums I’ve released so far. First of all, I have to let you know I use a simple audio interface with an XLR microphone jack that connects to my desktop computer via Firewire, and then I use a multi-track audio editing software application to mix and master on my computer. I have two microphones, and two acoustic guitars, one acoustic bass, and some drums and percussion instruments, a marimba, and some harmonicas. To see the exact brands and models of my equipment and instruments, you can check out all the detailed info and even pictures of them on the gear page of my website: www.scottcooley.com/gear. My recording process is to go as quickly as I can from writing a song to having a finished recording. I find that the less time I spend on trying to get it perfect, the better it sounds and the more I like the experience. I can go from starting to write the song to having the finished recording of it in about as little as one hour, and the most I ever spend would be about three hours on one song, start to finish – assuming the song is already written first. Sometimes I write a song in fifteen minutes, while others linger as drafts for years, but that’s outside the scope of this blog topic. I never memorize my own songs. This may surprise some people, but since I’m not a live performer, I don’t need to, and haven’t had the desire to. When I write a song, and when I get it somewhat complete, I record a first take on an old Sony recordable walkman cassette recorder. The first take cassette tape catalog and the finished digital recording are typically the only two recordings I have of most of my songs, and then I never play them again. Usually, that’s the only time I record it playing it live while singing it all the way through – on the 1st take cassette. I used to review the cassette versions and pick the best to record digitally, but now I just know which are good enough. Before I hit record on my computer though, I type up the lyrics (which I’ve usually done in advance of the time I decide to record when I wrote the song), and I also usually tune my guitar first.

Step One – Have The Lyrics In Front Of Me 

First, I need the song I wrote to be “written.” Years ago, I would hand-write w/ pen or pencil on paper the lyrics, sometimes writing the chord letter above the word where played, sometimes not. I’ve regretted not writing those chords down since the first-take cassette tapes wear out and listening back is the only way to know what the chords were. Since I have a bad memory, in recent years I try to remember to put the chord letters on the lyric sheets, which I now type in electronic document files and store on my computer. I used to print out the one-page song on my printer, and prop the sheet up, but now I even save that step because I have a wide computer monitor that allows me to tile vertically the screen. On the left is the lyric / chord file, and on the other the multitrack recording software.

Step Two – Record Rhythm Guitar Track 

The sequencer/editor/multitrack recording software application I use has a built-in metronome. I usually have an idea from when I wrote the song and played it through once on the 1st-take cassette what the tempo is going to be, so I set it accordingly and put on the headphones. I also know from the lyric sheet how the song is arranged because I’ll type the intro and instrumental break chords as sections along with the verses, chorus, bridge, repeats, etc. sequentially on the page. So, I’ll usually just write something like “last two lines of chorus chords” for the intro, or “verse chords” for the intstrumental break, and instinctively know to move my eyes to those sections to get the chords and play the right number of measures or bars, even though I don’t really know what measures and bars are exactly. So, with the headphones on, I’ll hit record, wait ether 4 or 8 beats, then start playing the rhythm guitar track all the way through. I usually nail it in the first take, but sometimes screw up, hit delete, and start over. I like to nail each track live in one take, so consider it a personal challenge to play every part perfect live while recording. It’s also a big hassle time-wise to do any tricky punch-ins or splices or whatever those are called because I’ve never really been able to figure out all the features of the software for that, and I think it loses some authenticity that way in the overall sound. I don’t really know how to fingerpick, but sometimes I fake it to get a good sound using three fingers on my right hand. Sometimes I sort of strum with just one finger. Other times I’ll use a standard flat pick to strum with. It all depends on the style of the song. Quite often, my preferred and signature sound is such that I like to have two rhythm acoustic guitar tracks and pan them out wide in the mix. I do not copy a single track and make one left and one right, however. Instead, I repeat the process and have two separetely-recorded takes. They never sound perfectly in sync with each other, and this way they really complement each other nicely. Sometimes if I knew I was sort of weak in a couple areas on the first take, I’ll accentuate them in the second. For both I use the AKG C1000S small diaphram condenser mic w/ phantom power about six inches away from the soundhole of my acoustic guitar, and I record it effect-free. Way later in my process I might add some reverb, but not much.

Step Three – Record Bass Guitar Track 

While listening to only the metronome in the headphones, I record the bass track. Rarely do I listen to the rhythm guitar track while recording bass, because it screws me up a little and makes the bass match the percussiveness of the guitar too much. I like it to be spontaneous and fresh, and metronome-only works best for bass for me. Again, I look at the lyric/chord sheet while recording. I use the Shure SM7B without the low cut or high boost, so the settings on the mic itself are flat, and then I crank the gain since it’s a dynamic mic. I play a non-upright acoustic bass guitar, and I cannot tell you how hard it is to get a good recording of this instrument, but the mic is the key. I’ve tried the Sure Beta 52 which is touted to be good for this, but I’m here to tell you it’s not at all – it sucks. Also early on I used the bronze wound strings, but since I switched to the black nylon, the sound is way better. It’s not as loud, but who cares when you’re recording, you can always boost the volume later. These strings allow me to avoid the sound of your fingers sliding across the strings, and it also makes it sound more like an upright bass. Additionally, you can avoid the miscellaneous pull-off and transitional clicks if you don’t play precisely enough. Some imperfection can sound cool, but too much is bad, so these strings seriously help. So, the mic and the strings are the key ingredients here. Sometimes I have to do two or three do-overs until I can nail it all the way through. I have no idea how to actually play bass, so I stick with mostly root notes. I’ll also usually do a little warm-up before I hit record, making sure I know what I’m going to play for each part, and practice the walking transitions between verse, chorus & bridge. My audio interface has built-in DSP effects, and for guitar I do Compression > EQ, but for bass I switch them to EQ> Compression in the signal chain. I read somewhere on the internet this is what to do, and it does sound better. I have no idea why, and don’t really care, but it works. Most people who record go direct with a plugged-in electric Fender bass, but I’ve never tried that. I know those are way easier to play, but you don’t get the neat acoustic sound I get.

Step Four – Record The Djembe 

Since I already have stuff set for bass, I always record djembe next because I use the same settings and microphone. I just angle it about 5 inches from the top. In the headphones, I’m listening to the rhythm guitar track only. I want it to be fresh and blend with the bass, so it doesn’t work well if I’m hearing the bass track, so I do rhythm guitar only in the cans. This I usually nail in one take, because I’m just playing it in a minimalist way, and the hits are where a kick drum would normally be. I hear in my head where I want the hits to be somehow, so instinctively know. The key to my success here is keeping it simple, and also, I’m slapping my knee quietly with the other hand. The free hand is slapping the snare drum part. That’s my secret. Instead of recording w/ overhead mics and playing djembe & snare on the same track, I have them separately panned in the mix in the final recording because I record them separately.

Step Five – Record The Snare 

I do this exactly the same as the djembe, only switch to the AKG condenser mic, and switch back to the Comp > EQ order on the input. So, I listen to rhythm guitar track w/ metronome in headphones, and this time I slap my knee with the free hand to what the djembe would be playing. I don’t want to be thrown off by the djembe sound, so I mute that in the phones, but my free hand slaps my knee where it would be played, while my other hand hits the snare – usually with a nylon brush. Again, I just instinctively know where I want the snare hits to be. The hard part is the “fills” or “transitions” or “runs” or whatever those are called in between verses and choruses. I usually figure those out and know what I want to do before I hit record. I don’t claim to be good or knowledgeable at any of this stuff, but this is how I produce the sound I get, be that good or bad. Technically, I know I’m “off” a lot, but I do the best I can and the authenticity is refreshing and the simplicity lets other parts of the song shine. It’s “light” and “complimentary” the way I play drums.

Step Six – Record Cymbal and/or Percussion 

Now I’ll do individual tracks using the same snare setup for hi-hat cymbal, crash cymbal, tambourine, shaker, whatever the song calls for. I have some conga/bongos, and a wood slit mallett drum. I’ve used a cowbell a couple times and a washboard too. I just sort of hear in my head what the song should or should not have and take it from there.

Step Seven – Record Scratch Lead Vocal 

Now I listen to only the rhythm guitar in the headphones and record the vocal. I get it close to how I wanted it and that’s it.

Step Eight – Record Background Vocals 

Now listening to the rhythm guitar and scratch lead vocal in the headphones, I’ll record usually four separate background vocal tracks. For this I switch to the Sure mic, only I use the low-cut filter and high boost switches on there. I usually nail these in 4 takes, then pan 2 left and 2 right for the mixdown later. Sometimes, if the octave/key is right for my voice I’ll do 2 high and 2 low and blend. Instinctively, I sort of “hear” where the oohs and aahs should go, if at all, and whether I should sing all or only part of a chorus. I usually plan this out before I hit record. If I can understand where I can hit different notes than the lead vocal I do, however, this harmony stuff is difficult for me to do naturally, so I usually go with a softer, laid-back, more breathy version of the lead vocal but hit the same notes. I’m always amazed at how good my background vocals sound given my limited ability, but this really seems to improve the overall sound of the recording for most songs.

Step Nine – Record Instrumental Solos and Fills 

I don’t do many “fills” throughout the song, but this is when I would do those, along with the solo intrumental break and intro parts, sometimes a melody line in the chorus. I might use harmonica or marimba, or slide guitar, but usually do a standard acoustic guitar solo using a pick. The hardest part is to learn the actual melody line if I’m not doing a free-form pentatonic jam. To do this, I listen back to the lead vocal track while playing the instrument to get the notes to match. I learn it just for the purposes of recording the track, and then never remember any of it. This would be just for marimba or slide or guitar. On harmonica, I just do a free-form thing that sounds close, since again, I have no idea how to actually play harmonica. My slide sound is a Hawaiian Weissenborn lap-style slide bar, so it has a nice acoustic sound I like. Having a true melody line sounds good on some songs, where on others, just wailing out a jam naturally sounds pretty cool.

Step Ten – RE-Record Lead Vocal 

Here’s where I try really hard for several takes, saving them all, then I go back and do a complicated listen/delete process where I remove the worst parts of each track, then eventually find the best parts of maybe four or even five takes and mute or silence out the rest and then bounce down a “greatest” of all five into a single mixed track, which serves as the final lead vocal. I know this is cheating, but I’m a really bad singer and this allows me to get it as good as I can.

Step Eleven – Muting/Noise Reduction/Effects 

Now I’ll go and mute unwated sounds, clicks, pops and use the software to perform noise reduction on every track. After each track is totally “clean” I’ll go through and add a little reverb to the lead vocal and sometimes the snare.

Step Twelve – Mixdown and Master 

The mixdown is just a software function, and then I’ll do a little EQ, Compression and and Normalization to the final mix. Maybe fade outs at the end of some songs. Then I trim to cut out the intro and end silence.

Overall 

When I assess my overall sound as a performer/producer/engineer, here’s the brutal honesty:

Acoustic-Only

My sound is all acoustic, which even if played and recorded perfectly, can still sound amateurish. If I were to use electric instruments, however, I would need to get a full drum kit and actually learn to play the drums better, which I don’t have the space or desire to do. Electric guitar and bass would make it sound more professional.

No Keyboard

I don’t have a piano, and the cheap casio I have doesn’t sound like a real piano, which is the sound I would want. My wife’s accordion on a few tunes sounds awesome, even though I have no idea how to record it properly. I suppose I could teach myself piano too, but don’t necessarily want to take on yet another instrument, particularly since that’s one where it’s more important to be able to read & write actual notation in order to understand how to play. Piano would make it sound more professional.

Weak Rhythm Section

Since I’m also self-taught in bass, drums & percussion, I’m just as bad as I am with guitar, probably worse because I’ve done it less. I would know enough to teach a band the general idea I have in mind, and if I were to ever put together a band, they could listen to my released demos and take it from there. Hiring session musicians to record on my demos to play these instruments would make it sound more professional.

No Effects

Effects-wise, I could do a lot more, but this would only sound good if I also played electric instruments to begin with and had a more polished overall sound. I like it to sound raw, unpracticed, and unpolished, so it is very real-sounding. More effects would mean a big cost – either for plug-ins, harwdware channel strips, pedals, etc., and a big learning curve to figure out how to use them. Effects would make it sound more professional.

Weak Vocals

Doing it more doesn’t necessarily mean improvement for me. Just as I admit I’m self-taught with instruments, I also have no formal vocal training. Not only that, but I don’t have any natural ability to control my voice. This is not something I have a desire to invest in to improve, either with money for lessons or time. I try to make the lyrics clearly understood, and muster enough style to get some emotion across. I do the best I can with what I’ve got, and it’s good enough for my purposes. I do envision my songs being sung by pros, and wonder how much better I’d think the songs were if that happened. Hiring professional vocalists for my demos would definitely make it sound more professional.

Mixing

Mixing to me is just a matter of getting the relative volume levels of each track right for how you envisioned the sound of the song, and also panning the instruments how you want them to be. The rest is just software. I pan acoustic rhythm guitars and background vocals out wide, drums less wide, bass and lead vocal in the middle, and that’s about it. I then try out how it sounds on a car stereo and readjust volume levels accordingly, which is technically part of mastering, not mixing I guess. I do go the extra mile with recording each track separately, so have more control even though it’s more time-consuming. I suppose a pro engineer would improve over what I come up with however.

Minimal Mastering

I don’t spend much time learning about mastering, but do the best I can with what I have available in the software. On the one hand I don’t want it to sound electronic and over-polished, but on the other I know I could do better. It’s just a matter of testing what all the options do to the sound, and I don’t want a room full of hardware racks and all that. Overall, I’ve found less is more, and I’ve learned that no amount of mastering can make bad tracks sound like a great song, so the performances of each track are way more important. For music that is this amateur-sounding, paying a pro doesn’t make much sense. Pro master would make it sound more professional though.

Conclusion

In a nutshell, it is what it is. I have lots of room for improvement, but limited capability to improve. I could invest more in better equipment and/or lessons, but don’t have the desire. I’ll never be cut out for being a great performer or singer. So, the pipe dream is selling a song, getting a cut. Honestly, I can’t help doing this, even if I never improve or take it to any other level. I can’t go long without writing songs, and recording the best versions of them I can. That said, I like to be ultra-efficient about the process. I’m someone with minimal talent/skill who maximizes it using a minimalist approach. My taste is such that I don’t want to use synthesizers and electronics to mask imperfections. I like to make it sound real, yet the best I can make it sound. I like the challenge to be able to play each instrument track perfectly live without any fancy tricks, but then I purge my temporary memory of how I was able to briefly learn all the parts and never play the song again. I know, it’s weird. I guess I wanted to have a huge catalog of songs first before I decided which ones were good enough to memorize and play live for people in the living room or at the beach fire. Then of course, I maintain the dream of having a famous artist record a version that becomes a hit, and also to put together a band where I’m the lead singer/ guitar player and play only my originals someday. The reality is I love doing it, and like to maximize my free time by going for a quality/expense/speed ratio that I’m comfortable with. Hopefully, reading this will help you do the same.

No comments:

Post a Comment

Comment away, there's no moderation here.