WAV Specs for Distribution: Sample Rate, Bit Depth & Format
For distribution, deliver a 16-bit or 24-bit WAV at 44.1 kHz in PCM format. Spotify wants WAVE_FORMAT_PCM, code 0x0001. Submit your native bit depth and don't upsample. 44.1 kHz is the streaming standard everywhere. 48 kHz is the video and broadcast rate.
This page is the file-spec corner of the mastering-for-streaming cluster. The loudness pages cover how loud to go and where to set your true peak. This one is narrower and more boring, which is exactly why people get it wrong: what bit depth, what sample rate, and what file format you hand to a distributor.
The short answer is a 44.1 kHz WAV at your native bit depth. The longer answer matters because a couple of small mistakes here (upsampling a 44.1 kHz master, dithering down to 16-bit when you didn't have to, sending a 48 kHz file built for video) quietly cost you quality or get a submission flagged. None of it is hard once someone lays out the real numbers.
I'll go through bit depth, sample rate, the WAV-versus-FLAC question, the Canada-versus-US distinction (there basically isn't one for streaming, and I'll show you why), and the one broadcast case where 48 kHz is the right call.
streaming delivery sample rate (CD standard)
preferred delivery bit depth, submit your native depth
the only WAV format code Spotify accepts (PCM)
video and broadcast sync rate
Key takeaways
- Deliver WAV at your native bit depth: 24-bit if you mastered at 24-bit, 16-bit if that's all you have. Don't dither 24-bit down to 16-bit before delivery. Let the distributor handle that conversion.
- 44.1 kHz is the streaming sample rate. If your master is already 44.1 kHz, nothing in the delivery chain resamples it.
- Never upsample. Apple says it plainly: a 24-bit/44.1 kHz master should not be pushed to 96 kHz. Upsampled files aren't equivalent to native high-res and are identifiable.
- Spotify requires WAV files in WAVE_FORMAT_PCM (code 0x0001). The WAVE_FORMAT_EXTENSIBLE variant (0xFFFE) is not accepted, and some DAWs export that by default.
- The file spec is identical for Canadian and US artists. 48 kHz only enters the picture for broadcast and sync delivery, where Canadian specs (CBC/Radio-Canada, following EBU R 128) call for it.
- Stereo only. Most distributors reject or flag mono submissions unless the track is intentionally mono.
What bit depth and sample rate should I deliver to a distributor?
A WAV at 24-bit, 44.1 kHz if your master was produced at 24-bit. If you only have a 16-bit master, deliver that. The rule isn't about highest numbers winning. It's about delivering what you made.
Distributors set a floor, not a recipe. The floor across DistroKid, TuneCore, CD Baby, and Velveteen is 16-bit/44.1 kHz WAV. Above that floor they'll happily take 24-bit, and several take higher sample rates. Your job is to clear the floor and then match your real master, nothing fancier.
| Minimum accepted | Recommended / preferred | |
|---|---|---|
| Spotify (direct) | 16-bit / 44.1 kHz | 24-bit / 44.1 kHz, prefers FLAC; WAV must be PCM (0x0001) |
| Apple Music (via Apple Digital Masters) | 24-bit / 44.1 kHz | 24-bit / 88.2 to 192 kHz, afclip + AURoundTripAAC validation |
| DistroKid | 16-bit / 44.1 kHz WAV | 24-bit / 44.1 to 96 kHz (32-bit and 192 kHz may be rejected) |
| TuneCore | 16-bit / 44.1 kHz WAV | up to 24-bit / 192 kHz WAV |
| CD Baby | 16-bit / 44.1 kHz WAV | WAV / AIFF / FLAC, must be stereo |
| Velveteen | 16-bit / 44.1 kHz WAV | 24-bit / 44.1 kHz WAV, follows distributor-level specs downstream |
Apple Digital Masters sets its floor higher, at 24-bit. That's the one program where a 16-bit master genuinely won't meet spec. Everywhere else, 16-bit clears the bar.
Why 24-bit if listeners can't hear the difference
Bit depth is just how many amplitude levels each sample can take. 16-bit gives you 65,536 levels; 24-bit gives you 16,777,216. More levels means a lower noise floor and more dynamic range to work with.
The numbers: 16-bit tops out around 96 dB of dynamic range, 24-bit around 144 dB. Nobody listens to 144 dB of range, and your earbuds can't reproduce it. The extra room isn't for the listener. It's headroom that keeps rounding errors from piling up while the file moves through processing and conversion. You deliver 24-bit because that's the resolution your master already lives at, and dropping it early throws away precision for no reason.
Don't dither down to 16-bit yourself
If your master is 24-bit, deliver it 24-bit. Don't convert it down to 16-bit before upload unless a distributor specifically demands it. The platform or distributor does that conversion with its own dithering when it needs to. Doing it yourself first just means the audio gets re-processed twice.
Sample rate: 44.1 kHz for streaming, 48 kHz for broadcast
Sample rate is how many snapshots per second the file holds. Nyquist's theorem says you need at least double your highest frequency to reproduce it cleanly. Human hearing tops out near 20 kHz, so 44.1 kHz (a 22.05 kHz ceiling with filter room above the audible band) covers the full range with margin.
44.1 kHz is the CD standard and the rate streaming platforms deliver at. If your master is already 44.1 kHz, nothing in the delivery chain resamples it. That's the cleanest path.
48 kHz is the video and broadcast standard. If your music lives inside a video project (film, a TV cue, a YouTube video with original audio), work at 48 kHz the whole way through, because converting at the tail end of a video post chain introduces phase problems. For music-only streaming distribution, 44.1 kHz is the right call.
The higher rates (88.2, 96, 176.4, 192 kHz) get accepted by Apple Digital Masters and a few distributors, but they buy a streaming listener nothing audible. Their real value is upstream, in the recording and mixing chain where higher rates make analog-domain processing more accurate. The delivery file doesn't need them.
Does the WAV spec change for Canadian artists?
For streaming, no. A Canadian artist uploading a 24-bit/44.1 kHz WAV to Velveteen, DistroKid, TuneCore, or CD Baby faces the exact same requirements as a US artist. There's no Canada-specific WAV variant, and there's no Canadian government loudness or file standard for music streaming. The targets are international industry conventions the platforms apply the same way everywhere.
The one place Canada-versus-US matters is broadcast. Canadian delivery to CBC/Radio-Canada follows EBU R 128 and requires 48 kHz audio. That's for sync placements and broadcast-destined masters, not the file you send a distributor for Spotify.
If you're chasing sync, keep two masters
For a Canadian artist pursuing film or TV placements, keep a 48 kHz broadcast master and a 44.1 kHz streaming master from the same mastering session. Don't convert a 44.1 kHz streaming file up to 48 kHz for sync after the fact. Maintain both from the start rather than resampling one into the other.
FACTOR's Juried Sound Recording program, if you're funded, asks for a final mastered recording at project completion but doesn't set a technical minimum beyond "it's a final master." In practice a 24-bit WAV is what's expected for a professional funded recording.
WAV vs FLAC vs AIFF: which format to send
WAV is the safe default. It's uncompressed PCM audio, accepted by every major distributor and platform. One catch worth knowing: Spotify requires WAV files in format code 0x0001, WAVE_FORMAT_PCM. The WAVE_FORMAT_EXTENSIBLE variant, code 0xFFFE, isn't accepted, and some DAWs export that without telling you. If a clean WAV gets rejected for a reason you can't explain, this is a prime suspect.
| What it is | When to use it | |
|---|---|---|
| WAV | Uncompressed PCM, the industry-standard delivery format | Default for any distributor; use PCM 0x0001, not the extensible variant |
| FLAC | Lossless compression, about 50% smaller with zero data loss | Spotify prefers it for lossless; not every distributor accepts it for submission |
| AIFF | Apple's uncompressed lossless equivalent of WAV | Fine anywhere WAV is, identical quality |
FLAC is mathematically lossless, so a FLAC and a WAV of the same master are bit-for-bit identical audio. FLAC is just smaller. Spotify openly prefers FLAC for its lossless tier. The hesitation is that not every third-party distributor takes FLAC for submission, so WAV remains the format that always works.
What you never do is deliver a master as MP3 or AAC. Those are lossy. The platform transcodes your file to its own lossy codec at playback anyway, so handing over an already-lossy file stacks two rounds of compression and you get generation loss. Master to lossless, deliver lossless, let the platform do the final lossy step once.
Stereo is the requirement for most distributors. Some flag or reject mono files outright, unless the track is genuinely meant to be mono.
Dolby Atmos is a separate delivery path
If you're delivering spatial audio, none of the stereo specs above apply directly. Dolby Atmos masters use the ADM BWF format (Audio Definition Model Broadcast Wave Format) and run on their own loudness target: -18 LUFS integrated with a -1 dBTP true-peak ceiling, per Dolby's spec. That's different from the -14 LUFS and -16 LUFS stereo streaming targets the rest of this cluster covers.
TuneCore and a handful of distributors support Atmos delivery to Apple Music specifically. It's a parallel submission alongside your stereo master. You still deliver a normal stereo WAV for all the platforms that don't support spatial.
Frequently asked questions
Can I just export an MP3 and let the platform handle it?+
No. Streaming platforms transcode whatever you give them into their own lossy codec for playback. If you hand them an MP3, that lossy file gets compressed a second time, and the artifacts stack. It's called generation loss. Deliver a lossless WAV or FLAC and let the platform do the single lossy conversion.
My master is 16-bit/44.1 kHz. Should I convert it to 24-bit before uploading?+
No. Padding a 16-bit file out to 24-bit adds no information. Deliver it as the 16-bit/44.1 kHz file it is. The only program that won't accept 16-bit is Apple Digital Masters, which sets a 24-bit floor. Every other distributor takes 16-bit fine.
What's the maximum file size or length distributors accept?+
It varies by distributor and I won't put a single number in their mouth without their page in front of me. A standard 24-bit/44.1 kHz stereo WAV of a normal song length is well within every major distributor's limits. Very long files or extremely high sample rates (192 kHz) are where you'd hit a ceiling. DistroKid for instance may reject 192 kHz outright. Check the distributor's upload page if you're delivering something unusually long or high-res.
Does the file format affect my LUFS or true peak readings?+
The format itself doesn't change the measured loudness of identical audio. A WAV and a FLAC of the same master read the same LUFS and the same true peak. Where format matters is the final lossy transcode the platform applies: that step can push inter-sample peaks higher, which is exactly why the true-peak guide in this cluster recommends leaving -1 dBTP of headroom regardless of your delivery format.
Should I add metadata like ISRC and artist name into the WAV file itself?+
For distribution, what matters is the distributor's submission form, not tags baked into the audio file. The ISRC, titles, contributors, and credits reach the DSPs through those fields. Fill the platform's fields carefully. That's what the stores read.
Get better release strategy in your inbox
Release planning checklists, royalty explainers, and artist strategy notes from Velveteen. No daily noise.
Was this useful? Send a signal or flag a correction.
Keep reading
Pillar guide
Mastering for Streaming
LUFS targets, loudness normalization, true peak limits, and file specs for every major DSP, written by a working producer.
Related guide
True Peak Limits
How true peak limiting works, why lossy codecs can push intersample peaks above 0 dBFS, and why -1 dBTP is the right ceiling for streaming delivery.
Related guide
Spotify LUFS Target
Why Spotify's -14 LUFS normalization level changes how you should master, and what to deliver to stop punishing your loudest tracks.
Related guide
Audio file prep
What file to hand your distributor: lossless WAV or FLAC, 44.1 kHz or higher, 24-bit, stereo, with headroom. Why you shouldn't master to a platform LUFS target.
Free tool · no signup
Check your metadata before your distributor does
Run your titles, credits, copyright lines, and ISRC and UPC codes through the free checker and catch the rejection-bait errors before you upload. It all runs in your browser.