Last updated: 04 February 2009
Published in:
Digitising analogue media |
Creating new digital media
Tags:
audio editing |
business & community engagement
In the physical world we learn how to interpret sounds through hearing and perception. In the digital realm sound waves are traditionally represented in graphical form. This document introduces some of the precepts helpful when beginning to work with digital audio in audio software packages.
In day-to-day life we instinctively react to the sounds we hear through how our brain interprets the information passed through our ears.
Analogue sound production relies on this aural perception and in production we intuitively apply changes to what we hear through the physical adjusting of dials, knobs and switches. Working with digital audio, and specifically digital editing, adds a whole new visual dimension as to how sounds are interpreted in a way that they are ‘read’. With practice, we can base what we see in a digital representation of sound (a waveform), into what we expect to hear.
A digital waveform of a voice speaking the word ‘hello’ in a digital audio editor will look something like this.
Diagram 1

A waveform displays a recording of the word ‘hello’
The up and down of the wave is the varying amplitude which moves from left to right in the time domain. Section A is the sound of the ‘h’ in hello. It is the quietest part of the word, hence the spikes are not so tall, and the sound has a high pitch (try saying the word to yourself), as a result the cycles of the wave are closer together. Section B is the ‘el’ sound of the word, which has roughly the same cycle length (frequency) throughout but has a rise and fall in volume during the first third of the section. Section C is the ‘lo’ sound of the word as can be seen to be getting gradually quieter from the length of the spikes until the end of the word.
Visually we can divide the wave into three sections, as shown in diagram 1.
Probably the most helpful skill you will start to pick up when you are editing audio is the ability to ‘read’ a waveform. This will come easier with experience of editing and will ultimately allow you predict the qualities of a sound based on what you can see. It will also help you to identify locations within files from what you can hear which helps speeds up the editing process.
As you begin to edit you will realise that the need to zoom in and out of waveform views, quickly and often, is essential when conducting editing tasks. All DAEs (Digital Audio Editors) will have a zoom in and a zoom out function which is often combined with a keyboard shortcut to make zooming even easier.
As already mentioned, the zooming tools are essential to the editing workflow and you will find you will use them more and more as your experience grows. The simple reason being that sound is made from samples taken at an incredibly quick rate and therefore to view individual samples and other minor instances within a sound file, a very high resolution zooming function is necessary. Zooming is generally relative over the x and y axis’, however some DAEs provide a function for zooming on each individually and both together.
It can be easy to feel lost and become confused as you zoom into audio waveforms as their overall shape and characteristics seem to change the closer you zoom in.
To help with this confusion it is important to have a reference to your location in the time domain, using the timeline as a visual locator can help you remember where you are when zooming in and out. Placing it on the location you wish to zoom in on beforehand can act as a good reference to where you are when working up close.
Auditioning (listening to) very small areas of audio when zoomed up close is often impossible to help you edit, as the length of the sound is so minute. This forces you rely on your eyes as to what the sound is actually doing. These are a couple of the difficulties you will come across when you begin editing audio, which as you edit more you will start to intuitively overcome. A patient head and a method of trial and error are likely to be your best assets as you begin to edit audio files.
The best way to understand any small edits you are making is to listen to the edited region in the full context of the file. You may find that what you are working on actually has no effect on the overall perceived sound. As a result, regular zooming out and listening to the bigger picture may help a lot of teething problems you may happen to have.
You probably, by now, understand the concept of the ‘digital audio file’. This is the file which contains a waveform and can be imported into a DAE for editing and so on.
Within DAEs audio files are treated as regions of audio irrespective of the file which contains the audio. This is because when complex editing and arranging takes place, it may be the case that one audio file has been separated into many sections, which are then moved around and subsequently bare no resemblance to the original audio file. Therefore it is termed that sections of audio are treated as individual regions within a DAE.
There are two methods of editing audio files within a DAE. The first method saves any alterations to a region by processing the waveform within the audio file it is contained in. This is called destructive editing as you are constantly changing (destructing) the waveform as you edit.
The other method of editing is not to make and save changes to the waveform but merely to save the information about the changes instead. This method is known as non-destructive editing as the original waveform remains unchanged but is processed with the information saved.
The key difference between the two methods of editing is that unlike non-destructive editing, changes cannot be later undone when conducting destructive editing.
It is always advisable to back up your audio files appropriately before editing within a DAE.
As with the zooming tools, there will be other tasks you will find repetitive and laborious when using the standard mouse control. Most, if not all, DAE’s will allow you to assign keyboard shortcuts for tasks to make editing and mixing more efficient.
These shortcuts allow just that, a quick way of performing a task that might normally require navigation through a menu or dialog box.
“Audacity(R) software is copyright (c) 1999-2008 Audacity Team.
Web site: http://audacity.sourceforge.net/. [It is free software
distributed under the terms of the GNU General Public License.]
The name Audacity(R) is a registered trademark of Dominic Mazzoni.”
Last updated: 04 February 2009
Published in:
Digitising analogue media |
Creating new digital media
Tags:
audio editing |
business & community engagement
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++