Video icon 64
Build your creative skills with practical video courses from Tuts+. Start your free trial today.
Advertisement

An Objective Look at Object Oriented Editing

by
This post is part of a series called Editing Audio: The Complete Guide.
How to Edit All Your Takes At Once in Logic Pro 9
Open Mic: Tell Us About Your Favorite Audio Editing Techniques

You have heard about it, nearly every DAW on the market today seems to have it, yet you haven't really tried it. I am of course talking about Object Oriented Editing, the audio engineers gift from the gods! It is object oriented editing that extends to us the ability to edit takes, adjust volume of a particular words, and adjust EQ or compression all dynamically without touching automation. And to top it all off? It is a non-destructive and non-linear process to boot!

So if you have never attempted any form of object oriented editing, why have you not tried it yet!? This tutorial will get you and your workflow into this modern mindset and help you create dynamic musical mixes easier and more efficiently. So if you are ready to make your audio engineering life a little easier, jump on in!


Conceptually Speaking

Before we can really begin to apply practical usage out of object oriented editing we first need to understand how it works. And before we do that we need to understand different DAWs have different levels of complexity and capabilities when it comes to object oriented editing. So before we move on lets first take a look at how object oriented editing effects the audio chain.

The power behind object oriented editing is that we are able to take an audio file and process just a small section of that one audio file in the same way we normally processes an entire track. This effectively creates two distinct levels of audio editing in our DAW, the normal track level editing and the object level editing.

It should also be noted that not all DAWs call these small chunks of audio objects. You will typically also see them referred to as clips, regions, items, etc. but the function remains the same.

After these two lower levels we then obviously move from track level editing to sub-mixes and ultimately master track processing, but it is this lowest level of editing and mixing that allows us quicker and more efficient control over a mix.

For those of you who are more visual learners, here is a simple visual to show the hierarchy of editing in a object oriented audio chain...


Pretty simple right? The trick with object oriented editing then becomes, what can my DAW do inside this chain? Ideally everything would be the answer but alas we cannot have it all.

Here is a quick rundown of some popular DAWs and their level of object oriented integration...

  • Pro Tools: As of version 10 Pro Tools has begun to add very limited object level editing called Clip Gain. This is nothing more than a fancy term for being able to change the volume at the object/clip level. Unfortunately Pro Tools as of version 10 does not have object based aux sends, effects, etc.


  • Logic: Unfortunately for Logic users, Logic tends to fall into the same boat as Pro Tools. While it does have clip based gain, fades, and automation via the region inspector, the object oriented controls stop there. You do have the option for region based automation and the hyper draw for midi regions, but unfortunately the object based effects, sends, etc. are not available.


  • Sonar: The Sonar users out there can rejoice as Sonar has one of the most fully fleshed out object oriented workflows out there. Sonar will refer to their objects as clips, but the automation, effects, gains, all work inside Sonar and allow for a truly dynamic workflow.


  • Cubase: While there is object based editing inside of Cubase it is a much more peculiar implementation. The object oriented editing in Cubase is strictly offline destructive processing which while it sounds bad comes with a bright silver lining; the offline processing history. This feature allows you do undo any of the destructive processing done to your audio objects even after you have saved and closed to the project which essentially creates the same basic workflow as a fully featured object oriented DAW but in an offline fashion.


  • Nuendo: Steinberg's other DAW Nuendo unfortunately falls into the same category as Pro Tools and Logic with only having gain on the object level. If you need to the object based effects, then Nuendo will probably not be your DAW of choice.

  • Digital Performer: A heartfelt sorry goes out to our MOTU friends who use Digital Performer as it has no object oriented capabilities in any regard. Hopefully they will get it soon!

  • Studio One v2: As of version 2, Studio One from Presonus is a fully capable Object Oriented DAW in the style of Cubase. Any processing done to an Event (Studio One's version of a object) is rendered, similar to a freeze track option, but can dynamically be undone at anytime as needed.


  • Reaper: The small and fast team over at Cockos with their DAW Reaper have an almost fully featured object oriented setup. You can add effects, adjust gain, pan, etc. via their Item FX option, but the only thing you cannot do is automate effects parameters and use object based aux sends. This does leave out some more nuanced techniques but should not be a deal for current Reaper users.


  • Ableton Live: Despite all of its features, Ableton lacks object oriented options. This is most likely due to its emphasis on live performance which obviously has no need for object oriented editing.

  • Samplitude/Sequoia: These two DAWs from Magix are by and large considered the most fleshed out object oriented DAWs available aside from Sonar. Everything from gain control, effects inserts, automation, pitch correction, etc. can all be done on the object level which allows for the greatest flexibility out there.


So the question now is should you keep reading if you do not have a DAW that is capable or fully capable of object oriented editing? Yes!

The reason why? Because you will be able to see the capabilities and workflow of object oriented editing and see if it appeals to you.

If it does, and I think it will, you can then look into other DAWs if you wish. Most of the DAWs listed at minimum have gain control on the object level which in and of itself is a huge efficiency boost.

For the remainder of the tutorial I will be using Samplitude as it has most all the options needed, but I will speak in conceptual terms so those ideas can be applied the DAWs above in some form or another.


Adjusting the Basics

Being a different system than most engineers are used to (especially in the US) we need to adjust how we think and approach editing and mixing with object oriented editing. Always do keep in mind the audio flow from the previous visual as you begin to adjust to an object oriented workflow. With that in mind lets first take a look at basic gain and panning.

When it comes to gain we traditionally adjust the overall level with the tracks, adjust our levels more finely with automation, then sends those tracks to a sub-mix to compensate for our automation as the automation probably destroyed our initial leveling, then send them either to another sub-mix or to the master.

What is wrong with this picture is that our automation is being used to correct issues with the recording, not enhance our mixing. How many times have you dreaded using automation because once you use it, it becomes a royal pain to change? Well with object oriented editing we can easily streamline this process.

  • First determine what needs gain correction in your track before it hits any effects as any changes done to the gain here will be pre-fx.

  • Next slice up your audio file into smaller objects so that you can individually change the gain on the problem spots. You will most likely find these on drum tracks or vocal tracks where a few words or hits got too loud or soft.

  • Adjust the gain off these problem objects so that you achieve a more balanced performance.

  • Now add your effects to the track and use the track fader to adjust overall level after any additional processing from compressors, reverbs, etc.


This now leaves your fader automation open for more musical applications such as slowly adjusting a snares volume in and out of the chorus, or perhaps the same thing with backup vocals. You can then use your sub-mix as a general leveler for the whole group of vocals, drums, etc. which is what it was intended to do!

The whole trick and point of this is to do our gain correction on the object level so that we can avoid needless automation that does not serve any musical function. This is also useful if you need to fix just one little section of a track and automating the fader for that one little section would be overkill.

Another idea with the gains to keep in mind is the use of automating the gain of the object and not the faders. If your track requires a lot of musical automation then by all means automate the fader itself, but if you need to automate just a simple volume change once or twice, why deal with the automating the fader when you can simply automate the gain on the object for that particular section.

Just like with the gain, we can use a similar approach to the panning as well. Why automate the pans for one small section and have to deal with correcting the automation for the rest of the song, when we can just adjust the pan on the object prior to the track?

One common usage of this technique would be when a vocalist sings a delay line themselves without a delay plugin because they wanted the inflection to change with each repeat. On top of that you wanted to pan the sung delays around to give it more movement.

You could of course automate the pans for that one section or you could just slice of up the sung delays and pan them all differently; easy ping pong effect.


Corrective Effects and other Possibilities

Aside from basic gain and panning properties, we can also use object oriented editing to help clean up our tracks from an EQ stand point.

Traditionally if a singer got too close to a mic and created a proximity boost, or if a guitarist accidentally bumped an extra string on a chord, we were left with a dilemma. Either rerecord the performance and try to avoid the mistake or painfully try to fix the mistake in the recording. And if the performance was perfect besides those little mishaps, well we get to fix it no matter what.

The usual way to fix this sort of problem would be to possibly duplicate the track and fix that one problem spot in the new track and try and blend it with the old track. Or perhaps automating an EQ plugin so that it only effected the track on that one small instance.

Either way, these approaches end up being lengthy pain staking processes that are not fun at all. Instead, using object oriented editing we can fix the problem in the same track without automation.

  • Find where the problem spot is in your track and slice it up so that problem area is its own object.

  • Next add the appropriate effect to the object in question. To correct too much bass or harshness in the object just an EQ; preferably a linear phase one since this is corrective situation. For noise or other extraneous sounds, try using a gate, denoiser, spectral analyzer, etc.

  • Adjust the plugin until the problem goes away just as you would using track level effects then playback a little before and after the object to make sure everything still blends together.


As you can see this is a much simple easier to use process for very common problems.

Another way in which the object oriented editing can assist us is with the use of aux sends. As you know aux tracks are often used for delays, reverbs, etc. and require sending a certain amount of our track to this aux track. Typically you see certain words, or certain drum hits, etc. sent to these aux tracks via automation because we only want those certain hits or words to have the additional effects.

And just like with any other kind of automation, we need to adjust the automation across the whole song so we don't accidentally get delays and reverbs when we do not want them. Instead, we could use object level aux sends in place of the automation.

  • Find a particular word, vocal, guitar riff, etc. that you think needs additional delays or reverbs added to it in order to make it stand out or add some musical motion to the song at that point.

  • Slice that particular part into its own object and send it via object level auxs to the aux channel instead of using the track level aux sends.

  • Go to your aux track and adjust the effects on that track to taste!


See how much quicker and easier that is? We completely avoided any automation and could instead spend more time adjusting the effect perfectly.

Is there a downside to this? As with everything yes there is. Earlier we were correctively editing our objects with EQ, gain, etc. and were focused more on pre-mixing editing. Here however with musical mixing effects such as delays and reverbs things may need to change later as the mix develops.

What happens if we forget we used an object aux send and not a regular track aux send? We will go nuts trying to figure out why we keep hearing a delay or reverb there when we think there should be none! The moral of the story here is to keep in mind where you put your effects and sends on objects or else you may go crazy later!


Conclusion for Now

So far we have analyzed jut what exactly object oriented editing is and how it can make our basic editing and mixing go by a lot quicker and with less hassle. To do keep in mind however where you start putting effects and the like on objects as they are not always easy to find if you forget.

With all of this covered, next time we can move into more creative uses of object oriented editing and how it can make creative decisions go by much quicker and allow for more musical freedom.

Until next time!

Advertisement