How To Export Stem, Scratch, And Demo Tracks (In-Depth)

Photo By Marcus Winkler at Unsplash

Let’s say you bought a cake. A nice, extravagant, large, beautifully detailed cake. You go to pick it up at the store you ordered from, and… it’s not what you pictured when you ordered it. Instead of it being a clean, bright white, it’s slightly yellow colored, uneven, and some dents/scratches in it. Later you realize that there was a "high quality" option before you hit order! After many head smacks you reorder it and it looks phenomenal. Well, the same can happen when you export tracks from your DAW!

It’s imperative to know what these settings are and what they do to deliver the best quality you can. Whenever you need to export stems (export individual tracks for mixing or editing), scratch tracks (tracks to be overwritten later but used to get a session going), or demos, you can refer back to this guide of what you should do when you need to send your files to someone else to work with, or even as the final file you’re going to release! Here’s a full guide of how to export stems, scratch tracks, and demos to make sure you have the best quality and the least confusion.

Export Settings

Getting the export settings right the first time will save you from doing it all over again, so make sure you ask your engineer what format of files they want! There’s 3-4 main points when it comes to quality and type of audio files:

• File Type (.wav, .mp3, AAC, etc)

• Sample Rate

• Bit Depth

• Bitrate (can be found from sample rate and bit depth)

It’s important to know what each file type is so you know what to give your engineer:

• WAV files are uncompressed audio files that inherently hold the most information (thus quality and large file size).

• MP3’s are compressed, meaning some information is thrown out in order to save on file size.

• AAC files are MP3’s successor, and stands for Advanced Audio Coding; it is compressed like the MP3, but generally keeps the audio quality higher at the same amount of file size. It’s important to note AAC is sometimes not accepted in certain software!

Sample rate is how many samples are played per second. It’s like video frame rate; low frame rates give you choppy video, it looks like fast changing pictures. Higher frame rates make movements more smooth, and you can more easily see and track faster moving objects; the same goes for audio.

Higher frequencies are much more detailed with higher sample rates, because there are more samples to describe the shape of the wave. Lower frequencies don’t get as much benefit because less samples are needed to describe the changes in amplitude of the wave.

44.1kHz is a typical sample rate, meaning there are 44100 samples that are read and played each second. 48kHz is often used with video, because 24 frames per second is a standard frame rate, and 24 divides 48000 nice and evenly.

88.2kHz and 96kHz are also typical, but aren’t as often used as the previous; this is because the file sizes are larger, and it uses more computer resources to process when mixing when a lot of plugins are used.

Technically, 88.2kHz and 96kHz should be much better in quality than the 44.1kHz and 48kHz, however the real audible difference is super small and not justifiable for the increase in file size and processing power needed; in fact, many trained ears have a hard time picking out the two!

CD’s and most streaming platforms take 44.1kHz anyway, so there isn’t much benefit to recording at these higher rates.

One important thing to note is that you need to record at these higher sample rates before you can export with this “higher quality” benefit; If you export a session recorded at 44.1kHz and export it with a sample rate of 88.2kHz, you’re still at 44.1kHz quality.

The only benefit of doing this, called “upsampling”, is that plugins can have more data to do math to get a minuscule amount of higher quality out of it. BUT, it’s so much hassle for no real noticeable difference! Especially if it’s being downsampled to 44.1kHz at the end anyway.

Bit depth is sort of like sample rate, but determines the quality/resolution in the other direction; the amplitude of the signal. The higher the bit depth, the higher the amount of “slices” there are from the max volume, down to zero. The higher the bit depth, the “smoother” the changes in volume there are, and the more precise each sample can be. Just like sample rate is the resolution in time, bit depth is the resolution in volume.

Because the bit depth determines how precise the sample volumes are, this determines how much noise is in the track. Since each sample needs to be a certain value dictated by the resolution of values, each bit of data needs to be forced into the determined sample value closest to the “real” value of the sample.

It’s like if you stand on a scale, and you really weigh 5.4 lbs, but the resolution of the scale is only one pound, the scale sees your weight as 5 lbs. For all you know, you weigh 5 lbs! This causes noise and distortion to occur in the signal as it pertains to audio. The higher the bit depth = lower noise floor and less distortion of the signal.

24bit bit depth is pretty typical to work with while recording and processing the audio, because the amount of noise is low; however, CD’s have a bit depth of 16bit, so most likely it will be down converted. 16bit is also a standard for streaming platforms, but they are also opening up to 24bit!

I always record in 24bit and leave it like that until it needs to be 16bit at the end (exporting the master file for CD and whatnot), or keep it at 24bit the whole time if I can!

So what settings should you export your stems, scratch tracks, or demos with? Well, ask your engineer! I can’t think of an instance where they would want something other than a WAV file, so stick with that. MP3’s can work depending on the bitrate, but it isn’t ideal by any means and should be avoided whenever possible.

Beyond the WAV file, some may prefer to work at a 44.1kHz sample rate, others 48kHz, others 96kHz. It’s good to be on the same page before you start recording though, so you can record with the correct settings. If not, you may have to upsample which isn’t ideal, or you may have to downsample which just adds another step and can get confusing.

Most all engineers will be using 24bit, occasionally 32bit-float. You should still be safe at 24bit quality-wise though if they do ask for 32bit-float after it’s too late and you’ve already recorded.

The bitrate is the amount of data that is processed in one second of audio. So, the higher the bitrate, the more information there is! This is more-so important for compressed file types like MP3’s, because this number basically represents how informationally compressed (not sonically) the audio is.

Track Layout And Organization

So after you’ve got your recording and export settings picked out, there’s still a few things you can do to clean things up in terms of organization and to avoid confusion from the engineer you’re sending your tracks to!

First up: clearly label your tracks. To keep them in order, you can add numbers in front like so:

1-Main Vocal

2-Guitar

3-Bass

Some engineers prefer different orders of tracks or naming systems so be sure to check with them on exactly how you should label your tracks!

To keep things clean, avoid the bunch of numbers that get added when making new tracks or playlists, like, “Main Vocal.dup.01.03”. If it’s a double/duplicate track, simply name it something recognizable and short, like “Main Vocal Double 1”, or if it’s a track that has the same part but panned on either side left or right, “Main Vocal Harmony 1 L”, L meaning left.

Labeling may seem super basic and annoying, but it can actually avoid some confusion and speed up organization for the other engineer!

Next up: Always include silence before the instrument comes in if the part doesn’t start right away so that all of the audio files start at the same spot. So if your tambourine only comes in at the last part of the song, you would have silence in the file from time zero to all the way until the start of when the tambourine comes in, and then end the file after the tambourine part is done!

This makes it so the engineer can simply drop all the tracks in without needing to find where they fit. This is super important for avoiding headache and speeding up the time it takes to get started!

Thirdly: Include a text document or something to send along with your tracks that includes the song’s tempo. This can allow the engineer to edit things to the grid if they are editing, and if they’re adding parts, they can stay in time by using a metronome!

Lastly: Include only the tracks you want the engineer to use. If you have 4 guitar takes but you only want to use one, just send him/her the one you want! If you want him/her to edit those takes into one, clearly label each by adding, “Take 1” or “Take 3” etc to the ends of the names!

Things To Keep In Mind When Exporting

Other than the export settings, there’s some things to keep in mind assuming you have everything labeled correctly and organized.

You’ll want to pay attention to the levels your tracks are at! If they are clipping and you export them, they will remain clipped and you can’t reverse it unless you export them again! On the contrary, you may want to boost the levels if your tracks are sitting very low in order to reduce the noise.

All you’ll need to look at is the output meter for each track you are exporting; if they are summed to another track and clipping before the summed track, and you have some processing on the summed track lowering the level to avoid clipping on the output of the summed track, you should be okay.

It is only the final output that matters when trying to avoid clipping inside your software (NOT analog gear!), assuming you like the way it sounds and isn’t clipping in a way you don’t like from plugins on the summed track.

---

I know this topic can get a little dry, but once you know it, it’s easy breezy! To simplify the process of preparing your tracks to be mixed or mastered, I’ve created a free checklist over at the contact page of the Fretless Studios site that you can follow along to make sure everything is good to go and the process can run super smoothly when sending your tracks to any engineer!

-Michael