Audio editing

I’ve discovered that audio editing takes an enormous amount of time. Not just in terms of the learning curve involved in getting acquainted with new software, but in the sheer time and concentration required for listening and relistening.

In terms of choosing which audio editing software to use, I settled very quickly on Audacity 1.2.6. It was recommended at the NPPA Summit, and it was free. I also had a 2004 version of the iLife’s Garage Band – and that may come into play in the future – but the obvious place to begin learning about audio editing was Audacity.

Along with the software and a plug-in necessary for dealing with MP3 files, I downloaded the online manual and every tutorial I could find, then printed them off for a “text” book nearly an inch thick.

Since I already had digital recordings of various presentations from the multimedia seminar, this seemed like good material to help me learn the basics of audio editing. I wanted to convert the lectures into brief sets of audio notes that could be “reviewed” quickly to keep myself enthused and focused.

Almost immediately my aging G4 Mac announced that its “startup disk” was nearly full and I should do something about that. Audio editing, it seemed, required a lot of memory, not to mention adequate processor speed. How would my Mac react when I finally got ready to put audio together with visuals? How would my wife react if I announced that I’d have to invest in more RAM, a bigger hard drive, and perhaps even a new Mac?

Well into the night and my shortest lecture file, it became clear that I also needed to develop a systematic approach to an audio editing project.

First and foremost, I needed to SAVE my work frequently to avoid having a lot of work wiped out by a disastrous software crash. (Audacity is reported to be very stable, so I presumed the occasional crash was due to the inadequacies of my computer.)

As for an overall editing strategy, I divided the task into three parts: 1) a rapid rough cut to select the parts I liked, then 2) a trim down session to tighten the selections as much as possible, and finally, 3) the fine tuning edit to even out the audio quality and smooth transitions between cuts.

Without this basic strategy, I spent a couple of hours boiling down a 30-minute presentation by Seth Gitner on “How to get started in multimedia.” The audio track revealed two immediate problems with the tiny Olympus WS-300 recorder: generally uneven levels of recording throughout the talk, and little patches of dead silence between the audible parts. If the speaker paused, the level of recording seemed to drop out entirely, then over-adjust its volume level when the speaker resumed, occasionally also dipping the volume with the leveling adjustment. The little silent gaps were easy enough to remove with the Select and Cut procedure, but I had to "Normalize" each swing and dip in the volume level of the recorded audio in order to obtain a consistent level for playback.

At the end of the evening, though, I had less than three minutes of audio notes – a nice, tightly-edited summary of what I wanted to remember from Gitner's presentation – in his own voice.

The trade-off was an uneasy feeling that what I wound up with probably could not be termed “documentary audio journalism.” It sounded “of one piece,” but was not. Most of the natural pauses for breathing were removed, as were inevitable pauses for thought, and initial sentences broken mid-sentence to be rephrased.

I even shifted some parts of his talk out of sequence, perhaps even out of context, in order to make more sense to my own ear.

Clearly, software for audio editing offered as many opportunities for manipulative abuse as software for photo editing.

The basic rule for the use of photo-editing software has been based upon prior experience: Do nothing to an image that you would not normally have done in the darkroom.

I had no prior experience editing audio. Where were the lines that should not be crossed?