Tips & Tutorials | September 24, 2014
Repairing audio involves carefully patching over troublesome areas and performing precise audio edits—all without causing audible artifacts that listeners may detect.
In this blog post, we’ll walk you through some basic repair and restoration scenarios and how to fix common audio issues using RX, our award-winning audio repair and enhancement toolkit.
Audio repair might be performed for an obvious benefit, such as salvaging a live recording that suffers from an audio dropout or sudden and distracting audio event. For example, removing springtime birdsongs from an audio recording intended to be set in the depths of winter, eliminating vocal plosives, and smoothing abrupt transitions in certain frequency areas are all ways in which audio can be polished in the post-production stage.
However, the principles of audio repair can also be used for more subtle edits and enhancements. That’s why it’s beneficial to establish your aim before setting out: Do you want to treat each file with individual care and attention, or would you rather define the most appropriate settings with which to batch process hundreds of files quickly?
The goal of good audio repair and restoration is to render the best possible sonic result with the least audible human intrusion. In essence, your intervention in the original recording should be transparent and not introduce new artifacts that distract the listener. Sometimes it’s possible to solve an audio problem entirely, and other times it’s about finding the right balance between reducing the problem and preserving the original audio.
Whatever your tastes, and even as times and nostalgic aesthetics change, the basic intention of restoration should remain the same: render the best possible sound with the least obvious interference.
No matter what software you choose, the audio repair process will require using some sort of visual editing tool(s).
Perhaps the most important is the spectrogram display, which is used to make precise selections of identifiable audio events. These precise selections can then be patched or repaired.
The tools that allow you to interact with the spectrogram are also important. Good audio repair and restoration software should include a number of selection tools, which you can use to draw or highlight specific audio events and frequencies that you observe on a spectrogram.
RX 4 TIP: In addition to the common horizontal and vertical time/frequency selection tools, RX 4 includes several advanced selection tools, such as Lasso, Brush, and Magic Wand. These provide much more control over selecting sudden audio events that change frequency and move about.
Terms such as brush and lasso are common across visual editing platforms. Here’s what they mean in the context of audio repair:
A lasso tool lets you use your mouse to outline a freeform selection of an image.
A brush tool lets you use your mouse to outline a freeform selection with a defined brush size. The brush size is usually adjustable.
A magic wand tool lets you automatically and intelligently select a specific audio event within a spectrogram (or certain pixels making up part of an image).
Once you’ve highlighted certain events, the audio processing that follows represents the final step of an audio repair.
Repairing and restoring audio typically involves working with the following types of processors:
Denoisers are used to reduce and remove steady state background noise. “Steady state” means noise that is not changing. It might include constant ambient noise or tape hiss (referred to as “broadband” or “noisy” noise), or electrical buzz and hum (referred to as “tonal” noise because it typically exhibits recognizable pitches or harmonics). Denoisers can be based on FFT with thousands of bands, or a simple crossover with just a few bands, and are sometimes designed for a specific use case, such as vocals.
RX 4 TIP: RX 4 includes a Denoiser that has adjustable thresholds in several bands with independent control for both tonal and noisy problems (it includes an envelope that may be used to shape the noise reduction curve). RX 4 also includes a Dialogue Denoiser, which is more suited for real-time noise reduction on spoken word or sung vocals within a mix.
Declickers are used to reduce and remove intrusive clicks and pops. These can be caused by anything from dust and scratches on an old record, a CD skipping on playback, or even mouth clicks and lip smacks from a voiceover.
Decracklers are closely related to Declickers, but are optimized to help reduce and remove a more continuous, quieter stream of clicks that blend together to cause what the human ear perceives as a general crackle.
TIP: Using a Decrackler before using a Denoiser is often a very effective way of dealing with surface noise recorded from vinyl or shellac records.
Declippers are used to repair digital and analog clipping artifacts. These artifacts occur when overloading an A/D converter or over-saturating magnetic tape.
Visual editing tools vary by manufacturer, but the basic premise combines visual representations of audio, via a waveform or a spectrogram, with tools allowing you to select and edit certain audio events rather than the entire file.
Dereverbs, or dereverberation processors, are a new, cutting-edge technology, and are designed to remove or reduce reverberations from audio. They are particularly useful for dialogue editing and ADR matching, and allow the engineer to remove unwanted or distracting reverberations from dialogue recordings.
RX 4 TIP: RX 4 Advanced Dereverb is one of several proprietary tools developed by iZotope.
With all of these tools available, you might wonder where to begin. There isn’t a single “correct” order in which to use them—it all depends on the audio material you’re restoring. Always begin with the most obvious or obnoxious audio problem that you can hear and identify. Then, depending on the audio, it may make sense to perform some processing tasks before others.
For example, a loud hum, a heavy crackle, or severe clipping might at first prevent you from hearing and dealing with additional audio problems. Peeling away that first layer may make the next step more obvious to you.
Don’t be afraid to try out different combinations of the tools to get the result you want!
Back up your work. Always make a backup of the original audio file before you begin attempting to restore it. Depending on the tool, some edits become permanent once the file is saved, so it’s always advisable to maintain a prior backup.
RX 4 TIP: RX 4 allows you to save your work and unlimited undo history as an RX 4 document, which can prevent losing or overwriting work.
Keep the ears rested and the mind open. While doing audio restoration work, you’ll likely spend a lot of time focusing on subtle details. Taking breaks will help you return with a fresh mind and see and hear the bigger picture.
Make multiple versions. Sometimes it helps to try doing the same audio repair more than once with different settings and then compare the results.
RX 4 TIP: RX 4 has a great Compare Settings tool that helps A/B results as you go. Also, you may come back to a version you tried a few days earlier when you were tired, and now find it sounding worse than ever. This happens to all of us! See suggestion #2.
Keep detailed notes. This is invaluable, particularly when there are so many different methods for dealing with different audio problems. For forensics work, documentation is often a required deliverable.
RX 4 TIP: Using RX 4 document files and saving module-specific presets can save you the trouble of writing out all of the parameters on a recall sheet as you would in the analog domain.
Back up your work—the first and last rule of any audio editing project! You never know when a hard drive, backup device, or original master might fail. Again, always back up your work!