Sunday, July 25, 2010

Progress of the Album: Editing (Ch. 13)

The new album continues to come along! So far, I’ve completed rough Pro Tools edits on seven of the twelve songs we tracked at Dark Horse Recording in Nashville on June 18th. Editing is definitely not the flashiest part of the production process, but it is crucial. In this blog, I will try to de-mystify this important stage.

Slicing and Splicing
Years ago in the days of reel-to-reel recording, audio editing was accomplished with a razor blade. If an audio technician wanted to make an edit in a recording, he/she would literally cut the tape where desired and re-attach the tape with an adhesive (usually tape). Now a days, editing has become much more easy and precise with the release of digital audio systems (the software Pro Tools being the most widely used). Today, audio can be sliced into pieces and re-positioned with a few simple clicks of the mouse – no razor or tape necessary.

Editing Basics
As I mentioned in the previous blog, I prefer to edit in layers (at least when numerous live tracks have been recorded). The first layer of editing involves two elements: (1) replacing wrong notes by cutting right notes from another take or another portion of the song and replacing the wrong notes with the right notes, and (2) swapping out any bass/guitar notes, drums fills, etc. which I deem as misplaced or inappropriate to a given song (this is pretty rare given the players at the June 18th Nashville session).

When opening the Pro Tools session of a song, I begin by setting up a rough mix so I can hear all the instruments. I start by turning down all the audio channels and then I carefully bring up the drum mics, starting with the overhead mics, followed by the kick drum mics, and then the snare drums mics (in addition, I'll mix in some tom mics and room mics here and there). I also bus all the drum channels to a single stereo audio track so that I can control the overall drum mix with a single fader. I then slowly bring up the bass and guitar channels until they are well balanced with the drums.

After I have a rough mix, I solo out the drums and listen to them very carefully from start to finish to make sure there are no questionable hits or fills. I will also re-adjust any edits that were made the day of recording. During the recording session in Nashville, we would sometimes re-record a portion of the song – this is otherwise known as a “punch-in.” For example, if we had re-recorded the first chorus of a song, I will listen carefully to the start and end of the first chorus to make sure the edits both at the beginning and at the end of the chorus are clean (no clicks or pops) and natural (undetectable by the listener). Usually a punch-in begins a measure or two before and lasts a measure or two after the section that is being re-recorded. This allows for some “handles,” allowing the editor to re-align the actual place of the edit if necessary. For example, I may find that the first edit sounds awkward if it happens right before the downbeat of the first chorus. I may instead re-position the edit to beat four immediately before the downbeat of the chorus. It all depends on the section being edited. Every edit is going to be different.

In most cases, I will add a cross-fade everywhere where there is an edit. A cross-fade connects two adjacent audio files by overlapping them with a quick fade out and quick fade in. Cross-fades help to eliminate any pops or clicks, and aid in smoothing out an otherwise abrupt edit.

The Second Layer of Editing
During the first stage/layer of editing, I do most of my editing manually. By this, I mean I cut and move audio around by ear. I am always working on a visual grid which can be set anywhere from whole notes to 1/64th notes and beyond. This visual grid can be helpful when I am trying to figure out why an instrument sounds ahead or behind the beat. I may also synchronize instruments by visually lining them up next to each (as I can see the actual audio waveforms). An example of this would be two unison acoustic guitar takes - one panned to left and one to right.

During the second stage/layer of editing, I use a software tool called Beat Detective that allows me to quickly tighten things up by a percentage. For example, I may edit the drums to 92% instead of 100% (which might happen if I manually lined up every drum hit to a 16th note grid). It’s a pretty amazing technology. Within minutes, the program can (1) find all the drum hits (2) slice the drum audio at every drum hit (3) align the drums to a grid by percentage, (4) fill any gaps caused by the moved audio, and (5) add cross-fades where necessary. This is much easier understood if you actually see the program at work. I may post a video tutorial one of these days.

As far as editing order, I usually begin by doing some moderate Beat Detective editing on the drums and then I re-adjust the bass and guitar as necessary. I also always keep a backup of the original unedited audio in case my edits end up hurting the groove of the song.

A Delicate Cascade
Editing is both miraculous and dangerous. You can save a song with editing, but you can also destroy it. It’s important to understand that an edit made to one instrument can greatly affect the feel of the other instruments in a song, and can quickly upset the rhythmic equilibrium. No edit is an island. Moving the bass can mess up the groove of the drums and guitars, and locking the high hat to a 16th note grid can make the drums sound robotic (which may be intentional, depending on the style and song).

Some styles such as dance music imply a robotic feel, but most rock music calls for a human touch with some moderate rhythm inconsistency. Even the most accomplished musicians can’t play with the precision of a drum machine, but this rhythmic “imperfection” is one key ingredient to making music sound alive.

So why edit? Well, editing becomes a necessary evil especially when a musician is combining live and programmed elements (which much of my music incorporates). Live drums have to be a lot tighter than usual if they are being played along side programmed drum loops and synthesizers. At the end of the day, it’s really a matter of taste, and I prefer my music to be tight, but not mechanical.

Most of the songs you hear on the radio these days have undergone significant clean up and editing. The real test always becomes whether a band/musician can reproduce this sound from stage (although the stage is much more forgiving than recorded media).

Pictured is a classic 2" tape machine. These machines still record great audio, but make editing very difficult. Some musicians will actually record to tape and then transfer the tape audio into Pro Tools where they can more easily tweak and edit the audio.

Monday, July 19, 2010

Production Update July 19, 2010

As you may have noticed, I have not been writing in the blog recently. The reason? I’ve been very busy in the studio. I’m currently going through the album song by song and am doing a variety of clean up. What do I mean by clean up? Well, for one, I am editing some of the drum, bass, and guitar audio we tracked on June 18th at Dark Horse Recording in Nashville. I usually go about editing in “layers.” This first layer of editing includes simple copying and pasting of parts. For example, if there is a wrong guitar note in one of the choruses, I will simply grab the right note from another chorus and paste it in where the wrong note was. This editing “layer” also includes some swapping of drum fills and bass licks. Occasionally, there may be a drum fill or bass lick that I’m not happy with, and I’ll grab a drum fill or bass note from another section of the song and paste it in where needed. All of this editing is being done in Pro Tools. Along with this process, I am also creating rough mixes of the live instruments for each of the songs. Once I am happy with the rough mix and the rough edits (editing layer 1, which I’ve been calling it), I will export a stereo mix (or bounce) of the drums, bass, and guitar and will import this mix of live instruments back into Logic (where the songs were originally created). I next delete all of the fake MIDI drums, bass, and guitars in the Logic session and replace them with the live instrument mix.

After this, I begin cleaning up all the programmed MIDI parts (in Logic). These programmed parts include piano (which will be replaced with real piano in the Fall), synthesizer, organ, electronic drum programming and effects. From time to time, there may be a wrong note, or I also may find that I need to either simplify or embellish some MIDI parts to now compliment the live instruments. I also go through a process of labeling every channel and instrument set up so I can easily rebuild the session down the road if need be. After the MIDI parts have been tweaked, I will begin exporting each individual instrument as an audio file. All of these audio files will later be imported into Pro Tools, as the album will ultimately be mixed in Pro Tools (not Logic).

In the coming months, I will be recording more acoustic guitar parts for the album with Matt Meyer (at 10x12 Productions). I will also begin layer 2 of the editing process, which incorporates a program called Beat Detective – I am still saving up to buy this. More about this later…

Sunday, July 11, 2010

Production Update July 11, 2010

The last several weeks since Nashville have been very busy, so I haven't had much time to work on the album until recently. This past week, I've been pulling rough mixes of the drums, bass, and guitar (from the Nashville session on June 18th) into Logic (they were recorded in Pro Tools). I've been focusing a lot recently on cleaning up the programming parts I've created in Logic (piano, synths, percussion, etc.). Also, yesterday, I tracked acoustic guitar with my friend Patrick Skelton for one of the songs on the new album. Patrick is also working on a new project, and we were also recording some of his guitar parts for his songs at 10x12 Productions.