This tutorial is about how to turn a long audio file into a folder of one-shot hits usable for audio productions. While it is applicable to sampling, it focuses on extracting drum-type hits from a junk percussion recording session. My hope is that a lot of people will see how easy it is to create custom sounds and make them a part of their productions. I'll be using Edison which comes with some editions of FL Studio, but also comes as a plugin for use in other DAWs. If you have the SliceX plugin you'll be able to follow along just as well. Also Included: Junk Percussion Pack with 23 sounds.
Step 1 - Setup
First I'll load my audio file into Edison. This tutorial will apply to drum sounds, beatboxed drums, sound effects, and anything percussive in nature. To make this file, I grabbed a bunch of pans, some coins, keys, and a spatula and started making sounds with them, and got about 12 minutes of audio from it. If you'd like to follow along with the steps I highly recommend that you go ahead and record some of your own hits and sounds, one or two minutes should be enough.
Rather than place the slices myself, I'll have Edison slice the hits for me. There are three options, and I chose Medium slicing, though "dull" slicing may have also worked. You'll want to choose the option that best fits your material, or you could place slices manually.
I listened on the playback and found that there is quite a bit of redundant material, so the first step for me is to narrow down the choices. Something you may want to do is disable Edison's undo for this part of the process. If you don't disable Edison's undo, Edison will save an undo audio file everytime you delete a hit, which will slow us down as we want to ruthlessly and quickly delete sounds, leaving only the best.
Another thing we can do right away to avoid clicks is force all slices to be on a zero-cross. A zero-cross is a point on a waveform when it intersects the zero value. When we select this tool, it adjusts any slices that aren't on silence to find the nearest zero point. Given the nature of this audio material, most of the slices should be placed right, but sometimes it doesn't work out that way, even with the zero-cross tool. And sometimes using the zero-cross tool will move a slice to a worse spot than before, so use it with caution.
The picture below highlights some zero-crosses.
Step 2 Choosing Sounds
To select the first slice, press CTRL+A to select all the audio in display, then press the left arrow key to select the first slice. From here, press the left and right arrow keys to preview the regions, pressing delete where necessary. At this point, don't worry about making adjustments or deleting the silence between hits too much. Also, keep in mind that these are going in your sound library and you don't want to fill it with lots of sounds, just quality sounds, so press the delete key on all the sounds that don't get your approval.
When deleting sounds, use your mouse to adjust the right edge of the selection. When you select a slice marker region, it selects the audio from that slice, all the way to the next slice. You don't want to delete the audio all the way up to the next slice, because sometimes Edison places the next slice marker a few milliseconds too late. So, if you delete an entire slice marker region, you run the risk of deleting a part of the next sound.
In Edison, when you delete a sound, you'll need to select a slice again, by right-clicking the marker name, to get back into the groove of pressing the right arrow to listen to the next slice, otherwise the playback will jump back to the beginning of the audio file, making you lose your place which can be frustrating. If you're using SliceX, on the other hand, you can right click the waveform display to select a slice, and by default you can use a midi keyboard to preview slices. These two workflow differences make it easier to use SliceX for this particular type of work.
Before Deleting (above). After Deleting (below)
After going through the sounds once, I decided to go through the sounds again, but this time I used a tool called "normalize all regions", which amplifies each region individually to the maximum volume. In audio, louder almost always sounds better to the ear, so by making everything about the same volume, you can make a fairer judgement of each sound.
After an hour break and another pass through, I've narrowed down the sounds from 200 to 43. When you have many sounds recorded any reason to delete a sound becomes a good one. Listening with headphones will help you pick apart the sounds in microscopic detail. With headphones I was able to pare down the number of sounds to 23 which is much more manageable. Now it's time for in-depth editing.
Step 3 Editing
One thing you will run into is extra noises, like in Slice 8 here:
You can also hear that there is a click at the end of the sample, and when this happens, it's important to check and see if the click is the result of a misplaced slice, because it might be the beginning part of the next hit. So let's check out slice 9.
It sounds fine the way it is, but let's move the marker just to see what it would sound like.
While it does sound interesting, I prefer the previous setting because it's more usable. So I'll use the undo button to return to the previous slice marker placement. While it's true that I disabled undo earlier, I can still do undos that don't affect the waveform.
I know I don't want that click at the beginning of Slice 9, and I'm certain I don't want it at the end of Slice 8, so I can safely delete it along with the rest of extra sounds in Slice 8. I'll select all the noise and press the delete key to get rid of it.
Before moving forward, I would like to show a case where adjusting the slice marker benefits the sound. If I play Slice 9 in full, you can hear there's a noise at the end, and you can see it in the waveform display. I suspect this noise might be a part of the next hit.
Let's take a look at Slice 10, and listen, before and after adjusting. Notice the difference a slight adjustment makes. It's subtle.
Step 4 - More Editing
You'll want to go through the audio file, adjusting markers where necessary and deleting extra noises that appears between slices. Something that will help you adjust markers is the "zoom on left of selection" option, as demonstrated below. Click on the Zoom button, which looks like a magnifying glass, to access it.
Step 5 - Finding the Noise Threshold
Now it's time to delete the silences between the sounds. First, you may want to reactivate Edison's undo or save the audio file, because we'll be making some significant changes when we trim the silence. Because there isn't absolute silence in our recording, we'll need to define "silence" for Edison. We want to define our noise threshold to be big enough to get rid of most of the silence, but we want it to be small enough so that we can preserve most of the tails of our hits. Just by looking at the audio file, find a hit that seems to have the longest tail.
We'll zoom in and select the area we'll consider silence. We'll be using a tool that will determine this threshold based on the volume of our selection. In Edison, we can resize the GUI vertically to help us determine that vague threshold where sound stops and silence begins. This technique may take some trial-and-error. So make sure Edison's Undo is enabled.
Under tools, select "Acquire Noise Threshold."
Another way that might be easier or harder for some would be to resize the GUI even larger and click-and-drag on the meter in the upper-right to set the noise threshold. A green band will appear in the GUI to show this. The parts that will be removed are the the parts that are quieter than the threshold.
Now that I have the threshold set, I'm going to go through each slice and delete the extra space between the shots. Select Tools > Trim Side noise to do this, as the pictures demonstrate.
If you find that this tool isn't working on a certain slice, it's a signal that the the slice does not end in silence, which might mean that the next slice needs to be adjusted using the "zoom on left of selection" technique shown earlier. Here's what the file looks like after using "trim side noise" on each slice.
As you can hear from the audio file, we now have a lot of great hits without silence between them that would work well together. I didn't do it on purpose but the sound file almost sounds like a usable percussion loop! From here, we should name the markers and export them individually. I will show you some shortcuts that make this the easiest part of the process.
Step 6 - Final Adjustments
"Regions > Rename All" will work great, because it will play the first marker, open a dialog for us to rename it, and it will move through all the markers like this automatically.
Now we're ready to export, but before we do, I want to suggest converting the stereo file to a mono file. The reason is that the left and right channels are the same, so it sounds exactly the same as a mono file, and a mono file will take half the hard drive space that a stereo file would. "Tools > Convert mono signal to mono format".
Lastly, it's time to export! Select "File > Export Regions As."
What you type here will become the prefix for all the audio files. Prefixes in audio files are useful for keeping track of where the audio file came from and if you drag these audio files into another folder, they will stay together visually because of the alphabetization. Now we have our one-shots, ready for use immediately!
Subscribe below and we’ll send you a weekly email summary of all new Music & Audio tutorials. Never miss out on learning about the next big thing.Update me weekly
Envato Tuts+ tutorials are translated into other languages by our community members—you can be involved too!Translate this post