Music Production File Formats: Stems, Bounces, and Masters

If you've ever sent tracks to a mixing engineer or tried to collaborate remotely, you've probably run into confusing terminology. "Send me stems." "I need a bounce." "Where's the master?"

And then there's the file format minefield. WAV? FLAC? MP3? 16-bit or 24-bit? 44.1 kHz or 48 kHz?

Here's the thing: getting this wrong doesn't just waste time. It can actually degrade your music or make collaboration impossible. So let's break it down in plain English.

What Are Stems?

Stems are individual groups of instruments exported from your project, all starting at the same time and running the same length.

Think of them like this: instead of sending someone your full mix (vocals + drums + bass + synths all mixed together), you send them separate files for each major element.

A typical stem package might include:

• Vocals (lead + harmonies combined)
• Drums (kick, snare, hats all mixed down)
• Bass
• Guitars
• Synths / Keys
• FX / Atmosphere

Notice these aren't individual tracks — they're pre-mixed groups. You're not sending 47 separate drum mic tracks. You're sending one "drums" file that has all your drum tracks already balanced.

Why? Because stems give a mixing or mastering engineer enough control to rebalance your song without drowning in hundreds of raw tracks. It's the sweet spot between "here's my final stereo mix" and "here's my entire project file."

Bounces vs. Masters

A bounce is just a stereo audio file exported from your DAW. That's it. You "bounce" your project to audio.

It could be a rough mix you send to a friend. It could be a demo. It could be a full mastered track. "Bounce" is the action, not the quality level.

A master is a specific type of bounce — it's the final, mastered version of your song, ready for distribution. This is what goes to Spotify, Apple Music, Bandcamp, etc.

Masters usually have:

• Proper loudness (around -14 LUFS for streaming)
• EQ and compression to make it sound polished
• Limiting to maximize volume without clipping
• Sample rate and bit depth matching distribution specs (usually 44.1 kHz / 16-bit for streaming)

So all masters are bounces, but not all bounces are masters. Make sense?

Which File Format Should You Use?

This is where things get messy. Let's cut through it.

For Stems:

Use WAV, 24-bit, at your project's sample rate (44.1 kHz or 48 kHz).

Why 24-bit? Because stems are "work-in-progress" files. An engineer might add EQ, compression, or effects. The extra bit depth gives them headroom to process without introducing noise or artifacts.

Why not MP3? Because MP3 is lossy — it permanently throws away audio data to save space. That's fine for listening, but not for professional mixing. You can't get that data back.

FLAC is also acceptable (it's lossless compression, like a ZIP file for audio), but WAV is more universally supported. If file size is a concern and you need to send stems over email, convert to FLAC temporarily. Just make sure the engineer can open FLAC files.

For Reference Mixes / Demo Bounces:

MP3 at 320 kbps is fine.

These are just for listening, not editing. MP3 keeps file sizes manageable for email or messaging apps. No one's going to master your reference mix, so don't stress about lossless formats here.

For Masters (Final Distribution):

WAV, 16-bit, 44.1 kHz for streaming platforms.

Most streaming services want 16-bit / 44.1 kHz WAV files. Some (like Apple Music) accept 24-bit, but 16-bit is the standard. Don't overthink it.

"But I recorded at 24-bit! Won't I lose quality?"

Not really. The dithering process (which your mastering engineer will apply) converts 24-bit to 16-bit cleanly. And honestly, 16-bit has a dynamic range of 96 dB — way more than you need for music. CD-quality is 16-bit for a reason.

If you're delivering to a mastering engineer, send 24-bit. If you're uploading to DistroKid or Spotify directly, send 16-bit.

Sample Rate: 44.1 kHz vs 48 kHz vs 96 kHz

Here's the honest truth: for music production, 44.1 kHz or 48 kHz is all you need.

44.1 kHz is the CD standard. 48 kHz is the video standard (if your track is going into a film or YouTube video, use 48 kHz).

96 kHz and higher? Overkill for 99% of projects. It eats up disk space, slows down your computer, and provides no audible benefit. Some engineers like the extra headroom for heavy processing, but most people can't hear the difference in a blind test.

Here's a simple rule: bounce at the same sample rate you recorded at. If your project is 44.1 kHz, export at 44.1 kHz. Don't upsample (e.g., exporting 44.1 kHz to 96 kHz). It won't magically add quality — it's like photocopying a photo at a higher resolution. The detail isn't there.

Common Mistakes to Avoid

1. Sending MP3 stems.
Just... don't. Ever. MP3 is for final listening, not for production work.

2. Normalizing your stems.
A lot of DAWs have a "normalize" option when exporting. Turn it off. Normalization changes the relative levels of your stems, which defeats the purpose. The engineer needs to hear your original balance.

3. Leaving effects on stems when the engineer asked for dry tracks.
If someone asks for stems, ask them if they want them with or without plugins. Usually, they want them dry (no reverb, no delay, no EQ) so they can add their own. But sometimes they want your creative effects. Just clarify first.

4. Forgetting to export from the start of the project.
All stems should start at bar 1 (or wherever the song begins) and run the full length, even if some tracks are silent at the beginning. This keeps everything aligned when the engineer imports them. Most DAWs have a "tail" option to capture reverb trails — turn that on.

5. Using random file names.
Don't name your stems "Audio_01.wav" or "Track 4.wav." Use descriptive names: "Drums.wav," "Lead_Vocal.wav," "Bass.wav." Your engineer will thank you.

File Conversion for Collaboration

Sometimes you'll need to convert formats quickly. Maybe you recorded at 48 kHz but the platform wants 44.1 kHz. Or you have FLAC stems but the engineer prefers WAV.

For that, a tool like KokoConvert's audio converter is handy. It handles WAV, FLAC, MP3, and other formats, and you can batch-convert entire folders at once. (No need to manually convert 12 stem files one by one.)

Just remember: never convert from lossy to lossless and expect quality to improve. Converting MP3 to WAV doesn't restore lost data. It just makes a bigger file with the same quality. The damage is already done.

Wrap-Up

Music production file formats don't have to be confusing. Here's the cheat sheet:

• Stems: WAV, 24-bit, match your project sample rate. Lossless always.
• Reference mixes: MP3 at 320 kbps is fine.
• Masters: WAV, 16-bit, 44.1 kHz for streaming (or follow the platform's specs).
• Sample rate: Record and bounce at 44.1 kHz or 48 kHz. Don't overthink it.

And if you ever get a confusing request from a collaborator, just ask them to clarify. Better to spend 30 seconds asking than to send the wrong files and have to redo it.