By Lampdog
Tue Dec 16, 2014 8:36 pm
20 Audio Editing Tips
You’ve finished your recording sessions, and now it’s time to process your audio.
Get on the fast track to sonic success with Hollin Jones’s essential advice…
1: Be Mindful of Snap Settings
With audio, as when editing anything in a DAW or a wave editor, keep an eye on whether snapping is switched on and what it is set to. If it’s set to a high value such as bar or ½ you will find precise edits almost impossible to make. On the other hand, if you are trying to cut a whole bar of audio you will want snap-to-bar switched on. As long as your audio is in time, using a snap value will help you make precise edits. A snap value of 1/16 or finer is usually helpful for working with transients inside an audio event. If you turn snapping off you get complete free rein to move events, but it’s also easy to accidentally de-sync your sound by doing this.
2: Zoom In
Software enables you to zoom in on any waveform to sample level, and this gives you a great deal of power. Glitches or other events that last for only a fraction of a second can be identified and processed or corrected by zooming right down. The finer the zoom level the quicker the playhead will disappear offscreen, so it can be a good idea to set up a loop and disable autoscrolling as this will probably drive you mad otherwise.
3: Use Spectrographic Editing
Regular waveforms are fine for seeing the amplitude and duration of audio files, but that’s about it. Spectrographic analysis, on the other hand, can show you multiple visual representations of the frequencies and amplitudes inside a sound based on the type of view that you select. Even better, they provide you with a way to edit sounds in whole new ways. Imagine you have a recording of the perfect vocal take but in the background there’s a car horn that’s crept into the recording – this would be tricky to remove using EQ because the frequencies of both sounds cross over to an extent. A spectrographic editor such as iZotope’s RX or Sony’s SpectraLayers Pro can show you this sound separately and you can then ‘paint’ it out, take a noise print or perform a number of other processes to ‘reach inside’ the sound file.
4: Process in Place
All serious software will let you apply audio processing to files, such as normalization, fades, reverse and usually also plug-ins as well. One interesting trick in some software is to process effects and other tools ‘in place’ on a file. This means opening the audio file in the sample editor, isolating a section of the waveform where you want to apply the processing and then ‘gluing’ it into the file. This doesn’t require you to cut the clip up first, and, for example, is a great way to insert some silence, to perform an EQ cut on one word or sound inside an existing clip, or to reverse a couple of notes from the middle of a guitar take. By processing inside the clip you can avoid cutting it up, though that approach is available to you too, of course.
5: Take Advantage of Compositing
Many DAWs support comping, which is the selection of multiple passes or versions of an audio clip and their combination into one, finished, perfect take. The usual way to achieve this is to record in a loop between the left and right locators, making sure your software is set to keep each take, mute it and record a new version with each pass. Then once you have stopped recording, go into the takes editor (your software will have a name for it) and audition each one. Maybe you’ll find the first line of take one sounded great, then the next line of take three, then the last bit of take one again. By cutting or marking each take appropriately you can build a perfect take from the constituent clips. Mini crossfaders are often available to smooth the transitions between takes.
6: Batch Process Audio Files
This is more applicable to situations where you have already made edits or set up plug-in chains and you want to apply the same settings to a number of files. Loading lots of audio tracks into a DAW, applying a track preset to each one and then exporting them as stems is possible but it’s a long-winded approach. Ideally what you want to do is batch process all your sounds at once. Software such as Sound Forge or WaveLab has built-in batch processing options. With these tools you can make all your settings for a single audio file, say, for example, a plug-in and EQ chain to clean up some voiceover recordings, and then apply it to a bunch of files at once. Hit ‘go’ and leave it to work. It’s a massive timesaver and you can even sometimes specify things such as auto fades at the start and end of each file, further saving you time.
7: Use Time- and Pitch-stretching
Any mid-level or better DAW will support time- and pitch-stretching of audio. They have different names: Flex Time and Pitch in Logic, Hitpoints in Cubase, and so on. Time-stretching is useful to make a clip fit the tempo of your project regardless of its original speed. Spend a little time properly editing its start and end points (since your DAW will work on clip length rather than waveform) and you should be able to snap it to a bar marker to fit it to the project. Alternatively, stretch audio without worrying about snapping to create some special effects, such as extreme slowdowns. Tempo-stretching can be done without affecting pitch, and the reverse is also true. Change the pitch of audio and you can conform it to your project key, duplicate a part to create harmonies and process a clip differently through your mixer.
8:Render Down and Duplicate Parts
When you work on digital audio, edits that you make are generally nondestructive, and that means you can usually go back to any step and undo it. Sometimes, though, certain kinds of edit are only possible on a real audio clip and not on one that’s being effected. Effects are generated in real time and therefore you couldn’t, for example, slice up a delayed clip because the software would analyse the source clip, not the sound of the delays since they were still virtual. The way around this is to simply bounce (not freeze) a copy of an audio part down either by exporting and re-importing it or by printing it to a new track internally. Then, any slice analysis is performed on the effected file, which will look very different to the original. Since you have copies of both you can keep the original too and decide which one to use for what purpose.
9: Use Special FX Plug-ins
You’ll probably be aware of regular plug-ins such as EQ, compressors and reverbs, but there are quite a few effects out there that are capable of much more extreme sound processing. There was a time when to get cut-up effects you basically had to physically cut up all your audio parts and process them through tons of effects. Now it’s much simpler with effects such as Turnado, BreakTweaker or Stutter Edit. These multi-effects simulate complex edits and processing, but instead of taking hours to work on they can be performed with a couple of clicks. You can take fairly ordinary-sounding
10: Extract Groove
Many DAWs will enable you to extract the groove from either a MIDI or an audio part, store this as a quantization preset and then apply this to another part. So you can impose your own groove maps onto recorded or sampled audio parts using this technique to change their feel. Software such as Melodyne and Cubase also lets you extract pitch data to MIDI; so, for example, you can analyse a vocal take and create a MIDI-triggered duplicate.
11: Cheat using copy and paste
Sometimes you will find that your recordings have some performance errors in them and there’s no opportunity to do any retakes to fix them, perhaps because the performer is no longer available. A good way to patch over such errors is to identify similar or identical passages in a take that were performed correctly and then isolate, cut out and copy and paste these into the location of the incorrect part. This takes more skill than you might think, since there may be small variations in timing or feel between the part you’re pasting in and the time segment you’re pasting it into. But with some careful nudging and perhaps even a little slicing and groove quantization of the audio, you can usually make it fit in, and if you do the job seamlessly nobody will ever know the difference.
12: Snap-to-Zero Crossings
Many DAWs have a ‘snap to zero crossings’ option that you can turn on when editing audio waveforms. This works independently of your main application snap settings and ensures that a cut is not made at a point in the waveform where signal exists, but rather where the level is zero. This means that when you edit two parts together you shouldn’t get any nasty clipping caused by clashing waveforms.
13: Use Volume and Fade Handles
You can automate a mixer easily enough but sometimes it’s quicker and more suitable to use the volume and fade handles found on audio clips in many DAWs. These work independently of the channel fader so you don’t affect any other clips in the track. Simply pull any chosen clip’s volume up or down, or create fades to create a sort of submix within the track.
14: Use REX Loops
Using software such as Propellerheads’ ReCycle or Reason you can slice up any audio file and quantize it as you like, even adding some effects before you bounce it out to a REX loop that can be opened in various different instruments and applications. These loops will alter their timing to fit any project and are also playable slice-by-slice from any MIDI input device. Audio is made super-flexible using this technique.
http://www.musictech.net/wp-content/uploads/2014/12/TEY1.jpg
15: Fix Glitches With the Pen Tool
In a sample editor you can zoom down to sample level, showing the very building blocks of the digital sound. A digital click or pop will usually manifest as a clear spike or peak in the sample display, sometimes flattening out against the very top. You can often fix these by using the pen tool to literally draw them out, flattening the waveform to erase the spike.
16: Always be Crossfading
When you edit audio clips and arrange them on the timeline, there can be occasions where two clips glitch as they cross over because one or both waveforms are not perfectly cut. To avoid this, apply short crossfades to remove any glitching. Some DAWs such as Cubase are able to automatically add these for you.
http://www.musictech.net/wp-content/uploads/2014/12/TEYY1.jpg
17: Learn a Few Shortcuts
Sometimes you might find yourself having to make lots of cuts to a long audio take, perhaps to remove periods of low-level noise between voiceover or guitar riffs. Rather than manually zooming and clicking over and over, use a shortcut or set up a macro to use a key command to split a clip. Over the course of many edits on a clip this will save you lots of time. Logic has a ‘strip silence’ tool to achieve this.
http://www.musictech.net/wp-content/uploads/2014/12/region-left.jpg
18: Use Regions
Instead of making physical cuts to audio files you can often define regions within a clip and then use markers or similar commands to jump back and forth between regions. If you decide you want to make edits permanent there is usually the option to divide clips to new clips based on the regions you have set up.
19: Remove the Room
One of the problems some people encounter when recording in less-than-perfect acoustic spaces is room ambience on recordings. Always use an isolator such as the Reflexion Filter, but you can also use EQ to try to identify and then dial down the frequencies that are clouding your sound. This takes time and patience but it can significantly improve the end result.
20: Know What Tools You Have
Each DAW will have a different (if overlapping) set of tools to make working with audio easier. These can save you lots of time, with tools for trimming, cropping, deleting overlaps and moving data to preset locations all available in many applications. These tools typically cut out tens of mouse clicks on your part, so if you’re spending hours on the donkey work of editing, look to see if you’re missing a trick.
You’ve finished your recording sessions, and now it’s time to process your audio.
Get on the fast track to sonic success with Hollin Jones’s essential advice…
1: Be Mindful of Snap Settings
With audio, as when editing anything in a DAW or a wave editor, keep an eye on whether snapping is switched on and what it is set to. If it’s set to a high value such as bar or ½ you will find precise edits almost impossible to make. On the other hand, if you are trying to cut a whole bar of audio you will want snap-to-bar switched on. As long as your audio is in time, using a snap value will help you make precise edits. A snap value of 1/16 or finer is usually helpful for working with transients inside an audio event. If you turn snapping off you get complete free rein to move events, but it’s also easy to accidentally de-sync your sound by doing this.
2: Zoom In
Software enables you to zoom in on any waveform to sample level, and this gives you a great deal of power. Glitches or other events that last for only a fraction of a second can be identified and processed or corrected by zooming right down. The finer the zoom level the quicker the playhead will disappear offscreen, so it can be a good idea to set up a loop and disable autoscrolling as this will probably drive you mad otherwise.
3: Use Spectrographic Editing
Regular waveforms are fine for seeing the amplitude and duration of audio files, but that’s about it. Spectrographic analysis, on the other hand, can show you multiple visual representations of the frequencies and amplitudes inside a sound based on the type of view that you select. Even better, they provide you with a way to edit sounds in whole new ways. Imagine you have a recording of the perfect vocal take but in the background there’s a car horn that’s crept into the recording – this would be tricky to remove using EQ because the frequencies of both sounds cross over to an extent. A spectrographic editor such as iZotope’s RX or Sony’s SpectraLayers Pro can show you this sound separately and you can then ‘paint’ it out, take a noise print or perform a number of other processes to ‘reach inside’ the sound file.
4: Process in Place
All serious software will let you apply audio processing to files, such as normalization, fades, reverse and usually also plug-ins as well. One interesting trick in some software is to process effects and other tools ‘in place’ on a file. This means opening the audio file in the sample editor, isolating a section of the waveform where you want to apply the processing and then ‘gluing’ it into the file. This doesn’t require you to cut the clip up first, and, for example, is a great way to insert some silence, to perform an EQ cut on one word or sound inside an existing clip, or to reverse a couple of notes from the middle of a guitar take. By processing inside the clip you can avoid cutting it up, though that approach is available to you too, of course.
5: Take Advantage of Compositing
Many DAWs support comping, which is the selection of multiple passes or versions of an audio clip and their combination into one, finished, perfect take. The usual way to achieve this is to record in a loop between the left and right locators, making sure your software is set to keep each take, mute it and record a new version with each pass. Then once you have stopped recording, go into the takes editor (your software will have a name for it) and audition each one. Maybe you’ll find the first line of take one sounded great, then the next line of take three, then the last bit of take one again. By cutting or marking each take appropriately you can build a perfect take from the constituent clips. Mini crossfaders are often available to smooth the transitions between takes.
6: Batch Process Audio Files
This is more applicable to situations where you have already made edits or set up plug-in chains and you want to apply the same settings to a number of files. Loading lots of audio tracks into a DAW, applying a track preset to each one and then exporting them as stems is possible but it’s a long-winded approach. Ideally what you want to do is batch process all your sounds at once. Software such as Sound Forge or WaveLab has built-in batch processing options. With these tools you can make all your settings for a single audio file, say, for example, a plug-in and EQ chain to clean up some voiceover recordings, and then apply it to a bunch of files at once. Hit ‘go’ and leave it to work. It’s a massive timesaver and you can even sometimes specify things such as auto fades at the start and end of each file, further saving you time.
7: Use Time- and Pitch-stretching
Any mid-level or better DAW will support time- and pitch-stretching of audio. They have different names: Flex Time and Pitch in Logic, Hitpoints in Cubase, and so on. Time-stretching is useful to make a clip fit the tempo of your project regardless of its original speed. Spend a little time properly editing its start and end points (since your DAW will work on clip length rather than waveform) and you should be able to snap it to a bar marker to fit it to the project. Alternatively, stretch audio without worrying about snapping to create some special effects, such as extreme slowdowns. Tempo-stretching can be done without affecting pitch, and the reverse is also true. Change the pitch of audio and you can conform it to your project key, duplicate a part to create harmonies and process a clip differently through your mixer.
8:Render Down and Duplicate Parts
When you work on digital audio, edits that you make are generally nondestructive, and that means you can usually go back to any step and undo it. Sometimes, though, certain kinds of edit are only possible on a real audio clip and not on one that’s being effected. Effects are generated in real time and therefore you couldn’t, for example, slice up a delayed clip because the software would analyse the source clip, not the sound of the delays since they were still virtual. The way around this is to simply bounce (not freeze) a copy of an audio part down either by exporting and re-importing it or by printing it to a new track internally. Then, any slice analysis is performed on the effected file, which will look very different to the original. Since you have copies of both you can keep the original too and decide which one to use for what purpose.
9: Use Special FX Plug-ins
You’ll probably be aware of regular plug-ins such as EQ, compressors and reverbs, but there are quite a few effects out there that are capable of much more extreme sound processing. There was a time when to get cut-up effects you basically had to physically cut up all your audio parts and process them through tons of effects. Now it’s much simpler with effects such as Turnado, BreakTweaker or Stutter Edit. These multi-effects simulate complex edits and processing, but instead of taking hours to work on they can be performed with a couple of clicks. You can take fairly ordinary-sounding
10: Extract Groove
Many DAWs will enable you to extract the groove from either a MIDI or an audio part, store this as a quantization preset and then apply this to another part. So you can impose your own groove maps onto recorded or sampled audio parts using this technique to change their feel. Software such as Melodyne and Cubase also lets you extract pitch data to MIDI; so, for example, you can analyse a vocal take and create a MIDI-triggered duplicate.
11: Cheat using copy and paste
Sometimes you will find that your recordings have some performance errors in them and there’s no opportunity to do any retakes to fix them, perhaps because the performer is no longer available. A good way to patch over such errors is to identify similar or identical passages in a take that were performed correctly and then isolate, cut out and copy and paste these into the location of the incorrect part. This takes more skill than you might think, since there may be small variations in timing or feel between the part you’re pasting in and the time segment you’re pasting it into. But with some careful nudging and perhaps even a little slicing and groove quantization of the audio, you can usually make it fit in, and if you do the job seamlessly nobody will ever know the difference.
12: Snap-to-Zero Crossings
Many DAWs have a ‘snap to zero crossings’ option that you can turn on when editing audio waveforms. This works independently of your main application snap settings and ensures that a cut is not made at a point in the waveform where signal exists, but rather where the level is zero. This means that when you edit two parts together you shouldn’t get any nasty clipping caused by clashing waveforms.
13: Use Volume and Fade Handles
You can automate a mixer easily enough but sometimes it’s quicker and more suitable to use the volume and fade handles found on audio clips in many DAWs. These work independently of the channel fader so you don’t affect any other clips in the track. Simply pull any chosen clip’s volume up or down, or create fades to create a sort of submix within the track.
14: Use REX Loops
Using software such as Propellerheads’ ReCycle or Reason you can slice up any audio file and quantize it as you like, even adding some effects before you bounce it out to a REX loop that can be opened in various different instruments and applications. These loops will alter their timing to fit any project and are also playable slice-by-slice from any MIDI input device. Audio is made super-flexible using this technique.
http://www.musictech.net/wp-content/uploads/2014/12/TEY1.jpg
15: Fix Glitches With the Pen Tool
In a sample editor you can zoom down to sample level, showing the very building blocks of the digital sound. A digital click or pop will usually manifest as a clear spike or peak in the sample display, sometimes flattening out against the very top. You can often fix these by using the pen tool to literally draw them out, flattening the waveform to erase the spike.
16: Always be Crossfading
When you edit audio clips and arrange them on the timeline, there can be occasions where two clips glitch as they cross over because one or both waveforms are not perfectly cut. To avoid this, apply short crossfades to remove any glitching. Some DAWs such as Cubase are able to automatically add these for you.
http://www.musictech.net/wp-content/uploads/2014/12/TEYY1.jpg
17: Learn a Few Shortcuts
Sometimes you might find yourself having to make lots of cuts to a long audio take, perhaps to remove periods of low-level noise between voiceover or guitar riffs. Rather than manually zooming and clicking over and over, use a shortcut or set up a macro to use a key command to split a clip. Over the course of many edits on a clip this will save you lots of time. Logic has a ‘strip silence’ tool to achieve this.
http://www.musictech.net/wp-content/uploads/2014/12/region-left.jpg
18: Use Regions
Instead of making physical cuts to audio files you can often define regions within a clip and then use markers or similar commands to jump back and forth between regions. If you decide you want to make edits permanent there is usually the option to divide clips to new clips based on the regions you have set up.
19: Remove the Room
One of the problems some people encounter when recording in less-than-perfect acoustic spaces is room ambience on recordings. Always use an isolator such as the Reflexion Filter, but you can also use EQ to try to identify and then dial down the frequencies that are clouding your sound. This takes time and patience but it can significantly improve the end result.
20: Know What Tools You Have
Each DAW will have a different (if overlapping) set of tools to make working with audio easier. These can save you lots of time, with tools for trimming, cropping, deleting overlaps and moving data to preset locations all available in many applications. These tools typically cut out tens of mouse clicks on your part, so if you’re spending hours on the donkey work of editing, look to see if you’re missing a trick.
Stop getting automatically offended when you see gold letters!