Backing vocals sound better when they're perfectly in time — and if yours aren't, Pro Tools can help you fix them.
You know the story: you have a great lead vocal, but the harmony or backing vocals just aren't in time with it, or the double for the chorus part just isn't tight enough. What do you do? The screen shows a typical problem, with the main vocal on the top track and three other parts below: as you can see, they aren't quite in sync.
The challenge: to make the three lower vocal recordings in time with the lead vocal at the top. In this month's Pro Tools workshop, we're going to look at four different techniques for sorting this out.
Manual Editing
This is what we did before we had any automated tools at our disposal, and it still has a lot to recommend it. With practice, it can be very quick, and it doesn't require any special tools or plug‑ins. In the screen above, I have cut up the first of the doubled tracks. The aim is to get the start and end points of each note tightly aligned with the lead vocal, as well as any obvious hard consonant sounds ('b', 'd', 'g', 'k', 't' and so on) that might distract from the rhythm of the track if they are not properly aligned. As you can see, this involves making a number of cuts, moving words and syllables about, and using crossfades to disguise the joins. In some places, for instance between the first and second words, I have filled gaps with a short section of 'atmos' — background noise from a quiet part of the recording — which is less obvious than complete silence.
This is the best I can achieve by manual editing. Note the sections of 'atmos' pasted in from elsewhere in the tracks to fill gaps. You'll need to make full use of Pro Tools' zoom function to ensure you're putting everything precisely into time, and listen repeatedly to the results.
Cut & Stretch
There are times when basic editing won't fix all your sync problems, especially when a note is too short. In the screen below, you can see that the note in the third region is a little short, and the long note in the fourth region is very short.
When notes are too short, time‑stretching may be preferable to editing. You can make a note longer by cutting it, moving and dragging the region boundaries — in effect repeating part of the note — but you usually can hear the edit points, and time‑stretching can yield more natural results. There are many ways of stretching a region, but I tend to use the TC/E trim tool, as part of the Smart tool.
One of the Trim tool modes is TC/E, which applies time‑stretching and also works with the Smart tool's Trim mode. Use the Smart tool in Trim mode to do the cutting, then change it to TC/E Trim and do the stretching. Don't forget that the TC/E Trim tool uses TC/E plug‑in and settings, which are set in the Processing pane of the Pro Tools Preferences window. If you don't have any additional time compression and expansion plug‑ins, you will be offered the choice of Digidesign TC/E or Time Shift plug‑ins. The Digidesign TC/E plug‑in is great for speech, but for anything pitched, it is worth using the Time Shift plug‑in. Once that has been selected, choose the most appropriate Settings file from the lower drop‑down menu. In this case I chose 'Female Vocal', because that is what I am working with. Remember that it's not a good idea to apply multiple time‑stretch operations to the same region, or you end up stretching or compressing already processed regions and the quality quickly degrades. If one time‑stretch doesn't quite work, hit undo and return to where you started, before trying it again.
In this case, it took several goes to get things right, and as you can see from the finished result
To put this long note properly in time, I've had to split it and stretch the two halves separately. (above), it turned out that simply time‑stretching the entire fourth note didn't work. Instead, I had to split it in two, and move and time‑stretch both sections independently, then crossfade the join. Note that you have to do your crossfading after your time‑stretching, as TC/E won't work if the region has a crossfade associated with it, so make sure you create an overlap that can support a crossfade.
Elastic Audio
Elastic Audio is an excellent way to sync up vocal parts. In the example I'm using, the track wasn't recorded to a click, so I don't have bar and beat marks to work to — but that doesn't mean I can't use Elastic Audio, and more specifically, the Individual Range Warp feature.
The first thing to do is turn Elastic Audio on for the relevant tracks. As this is a monophonic vocal line, I have selected Monophonic mode. When you switch Elastic Audio on for a track, the regions will go blank briefly as Pro Tools analyses the audio. To carry out the correction, I have put the reference track into Analysis mode, so I can see the individual event markers to help the process of lining up the other tracks visually.
I then put the track I want to sync to the reference track into Warp view. The first three markers on the lower track now become Warp markers, which appear as black lines with a little triangle at the bottom (see the screen below). Be careful not to move Event markers (grey lines) with the Grabber tool as you will apply the Telescopic Warp feature, which you don't want. (If you do this, use undo very quickly!) Instead, hold down the Shift key and click one of the Event markers with the Grabber tool, and you will enable the Individual Range Warp feature. Now the selected Event marker and the Event makers on both sides of it will be turned into Warp markers, and you can move the selected marker so that the start of the phrase lines up with the reference track.
Elastic Audio in action: when I drag one of the Warp markers on the lower track, it adjusts only the individual regions either side of it. Avoid moving Event markers (the grey lines to the right). Pro Tools will use the adjacent Warp markers as anchors, ensuring that no damage is inflicted outside the region you want to work on. Once the Event markers have turned into Warp markers, you don't need to hold down the Shift key. If Pro Tools creates any spurious Warp markers, you can delete them by right‑clicking on them and selecting Remove Warp Marker from the contextual menu that will appear.
I find the Monophonic Elastic Audio to be very good with vocal parts, producing minimal artifacts unless you push it too hard — in which case I would recommend getting the vocalist to re‑sing the part! You can use the X‑Form Rendered option from the Elastic Audio menu for higher quality; this will render a file each time you make changes, but you will have to wait while it does it, so I tend to work in Monophonic mode until I am happy, then render it using X‑Form at the end of the process.
Vocalign
Synchro Arts' time‑alignment plug‑in comes in a number of forms; I have the basic LE version that came with the DV Toolkit bundle for a while.
The automated option: Synchro Arts' Vocalign. The plug‑in was designed for use in the audio post‑production world to help tighten lip sync for recorded dialogue tracks, but it can be used in music applications for tightening up multiple vocal lines.
Highlight the region you want to be the reference track, open the plug‑in from the AudioSuite menu, click the Guide button on the left‑hand side, and then click on the Capture button. The top Waveform box will show the waveform for your reference track. Now repeat the process, highlighting the region you want to sync up to the reference track, then select the Dub track on the Vocalign plug‑in window and hit Capture again. Next, click the Align button on the left‑hand side of the plug‑in window, and a trace of the Dub profile will be superimposed on the reference waveform in the plug‑in window. Then select the destination track where you want Vocalign to put the realigned audio — normally I would put it back on the track where it came from. Finally, hit the Process button and watch Vocalign do its thing.
There you have it: four different ways of getting your vocal parts in sync. Enjoy!
