Audio Editing and Working with Loops


Topics This Week


Bounce to Disk

Recording a performance to disk is only one way of making an audio file. Once we've recorded, edited, and mixed a song in GarageBand or Reason, we'll want to make a mixed stereo file. We can take any of our recorded audio, or MIDI tracks that play back using virtual instruments, and render them to a new audio file in a process called bounce to disk. When we make a new file this way, we'll be taking the mixed and processed audio data from our master output and using it to create a new audio file.

Follow this link for a tutorial in bouncing a mix to disk in Reason.


Digital Audio Editing

Once audio data is recorded to disk, we can access any part of a recording instantly. This leads to some ways of editing that are unique to disk-based recording. There are three ways to edit audio once it's stored on disk:

Throughout this week's class we'll be looking at various ways to edit digital audio that will illustrate non-destructive and destructive editing.


Non-destructive Editing

One of the most common types of editing done with audio is rearranging the order of audio events. Much like we do with a word processor, digital editing allows us to copy and paste sections of audio to revise a musical performance. As we'll see, the word processor analogy only goes so far, since the written word and audio recordings are very different types of media.

Some of the things we can do with digital audio editing include:

The power of digital editing comes from the ability to define a region. A region has pointers to the beginning and end of a section of a file. Once defined, that section can be accessed at any time. These region definitions can be stored in a file and can be reused whenever, and as many times as we want. Since we're not changing any of the actual audio data in a file, this type of editing is non-destructive.

In GarageBand, we can define and edit regions in the Timeline. Editing audio regions in GarageBand is exactly the same process as MIDI regions. We can shorten or extend, separate, and repeat regions. As with MIDI regions, we can choose to have audio regions move freely or snap to a bars and beats grid.


Destructive Editing

Although the word "destructive" may sound ominous to you, all it really means when it comes to digital editing is that you're rewriting a file. Although non-destructive editing techniques are very powerful tools for manipulating musical ideas, there are some situations where we are better served by saving our edits as part of a file.

A file that's going to be used as a master for any kind of distribution should be a final, completed version. Most of the non-destructive edits we make need to be played in the software that created them in order to be heard. The regions we define in a GarageBand audio track won't be recognized by a CD or MP3 player, they need to be saved as an audio file in iTunes. If we want to edit them further, we'll use destructive editing techniques. In addition, we can process an audio file to permanently change its fundamental characteristics.

Here are some of the ways you'll use destructive editing and processing techniques:

Far from being "destructive" in the usual sense, these types of edits offer a number of creative possibilities to reinvent the sounds we use in a desktop production.


Editing in Audacity

Audacity is a free, open source audio editor for OSX. We can use it to complete some basic exercises in audio editing.

We can edit a song form using standard Cut, Copy, Paste, Delete, and Trim commands. Cut, paste, delete, and trim functions will change the length of a file. To remove audio without changing the length of a file, use the Silence command.

The commands found in the Effect menu allow us to process the audio data. Using these, we can perform the following functions:

We'll be looking at ways to process audio in real time in Week 11 when we discuss signal processing.


Audio Loops

Audio loops are recordings that are edited to fit exactly into a musical meter. Any audio editing program can be used to define loops from a source and save them as separate files. Typical loop lengths are usually groups of "duple" phrases -2 bars, 4 bars, 8 bars. Longer phrases will have more variation, although the tempo of a human performance might shift. A good loop strikes a balance between the two.

A Word About Loops

One of the very first digital audio tools available to musicians was a sampler. Although the original intent was to make new types of synthesizer sounds based on audio recordings, a number of enterprising producers began sampling short rhythmic sections of existing songs. These were often drum breaks -sections of a song where only the drums were playing. Using the looping functions found on any sampler, these break beats, as they're called, can be repeated and used as the rhythmic basis for new songs.

Needless to say, this use of copyrighted material has raised numerous legal issues and has stirred an ongoing debate on intellectual property. What constitutes fair use of copyrighted material? If you use two measures of a James Brown recording and create a new song from it, what do you owe James Brown? In response to these legal challenges, a cottage industry of soundware developers has sprung up, providing musicians with libraries of breakbeats, instrumental performances, and virtually any sound you can imagine, all cleared for use in a variety of productions. You'll find examples of these as DrREX loops in Reason or Apple Loops in GarageBand.

The main advantage of using these types of loops in a production is that it allows you, the producer, to incorporate a wide variety performances that you ordinarily might not have access to in your song. Many of these are high-quality recordings, done in a professional studio with well-known pro players. Although the subtlety and variation that comes with a great live performance can't really be captured in a few short sample loops, creative use of these tools can definitely boost the production values of your work.


Working with Loops

There are two ways to work with audio loops:

Rex Files

DrRex is a loop player in Reason that plays back REX audio files. These files are created with a program called ReCycle, which is a tool for analyzing transients and saving the result in a single stereo REX file. Since slices played at a slower tempo will result in audible gaps, ReCycle can generate additional audio at the end of each slice that serves as a kind of padding. Again, as with time-compression/expansion, there are limits to this. A 2-bar loop at 135BPM will have prominent gaps when played back at a slower 72 BPM.

A REX file stores:

Apple Loops

The Apple Loop file format is similar to a REX. However, in addition to defining audio as individual slices, pitched musical parts can be recognized by key. In a program like GarageBand, this information can be used to transpose audio performances that have been saved as Apple Loops.

An Apple Loop stores:


This Week's Recommended Assignment

- Audacity is a free audio editor for OSX. Use it to review the basic exercises in audio editing we covered in class.
- Try assigning audio files to Redrum instruments and make your own percussion/sound kit. Create a simple etude that incorporates it.


Review Questions
  • What is the difference between bouncing a mix to disk in GarageBand and Reason?
  • What kinds of GarageBand tracks are included in a bounced mix file?
  • What is non-destructive audio editing? Give an example of how it is used.
  • What is destructive audio editing? Give an example of how it is used.
  • What type of audio editing can be done in GarageBand?
  • What type of audio editing can be done in Audacity?
  • What is the difference between changing the level of an audio track in GarageBand and changing the level in Audacity?
  • What is the difference between a REX loop and an Audio Loop?

Digital Audio Editing Vocabulary

Apple Loops
Audio tempo
Beat slicing
Bounce to disk
Constructive editing

Destructive editing
Non-destructive editing
Region
REX files
Time compress/expand