Audizer
Loudness and mastering

Export settings for a finished song

For a master going to streaming: WAV, 24-bit, at the sample rate you recorded in. For a mix going to mastering: the same, or 32-bit float. For CD: 16-bit, 44.1 kHz, with dither. Don't upsample, and don't export to MP3 for anything but a quick reference.

What to choose

DestinationFormatBit depthSample rateDither
Streaming distributorWAV24-bitSession rate (44.1 or 48 kHz)No
Mastering engineerWAV24-bit or 32-bit floatSession rateNo
Stems for a collaboratorWAV24-bit or 32-bit floatSession rateNo
CDWAV16-bit44.1 kHzYes
VideoWAV24-bit48 kHzNo
Reference for a phoneMP3 or AAC, 256 to 320 kbps

Most distributors accept 16-bit or 24-bit WAV at 44.1 kHz and above. Check your distributor's current upload requirements before the first release.

Bit depth

Bit depth sets the noise floor of the file. Each bit is worth about 6 dB.

Bit depthDynamic rangeNotes
16-bit96 dBFine for final delivery. Needs dither when reducing from higher
24-bit144 dBFar more than any converter or room can deliver. The standard
32-bit floatEffectively unlimitedValues above 0 dBFS are stored and can be turned down later without clipping

24-bit is enough for any delivered master. 32-bit float is useful between stages, because an accidental over isn't destructive.

Sample rate

A sample rate captures frequencies up to half its value. 44.1 kHz covers 22.05 kHz, past the limit of hearing. 48 kHz is the video standard and a common recording rate. Higher rates give plugins more room and make files twice the size.

Export at the rate the session ran at. Sample-rate conversion is a filter, and a careless one adds artefacts and can raise peaks. If a destination needs a different rate, convert once, with a good converter, at the end.

Exporting a 44.1 kHz session at 96 kHz adds no information.

Dither

Reducing bit depth throws away the lowest bits. On quiet material that leaves a faint, grainy distortion tied to the signal. Dither adds a tiny amount of noise before the reduction, which turns that distortion into a steady hiss far below the music.

  • 24-bit or 32-bit float to 16-bit: dither. Once, as the very last step.
  • 32-bit float to 24-bit: optional. The effect sits around −140 dB.
  • Staying at the same depth: none.

Other settings

Normalise on export: off. You set the level with the limiter.

Real-time or offline: offline is fine unless you use hardware inserts.

Tail: include enough time after the last note for reverb and delay to finish. 2 to 4 seconds is typical.

Start: begin a few hundred milliseconds before the first sound, so no player clips the first transient.

Mono or stereo: stereo interleaved for a master.

Check the file

Open the exported WAV in a fresh session or a player and listen to the whole thing. Confirm the true peak and integrated loudness of the file match what you measured in the session. Problems at the start, end or at edits show up here and nowhere else.

Quick answers

Should I export at 16-bit or 24-bit?
24-bit for streaming distribution and for sending to mastering. 16-bit only when a destination requires it, such as CD.
Should I export at 44.1 or 48 kHz?
Whatever the session was recorded at. Converting adds a processing step and no quality.

Audizer does this when you ask.A Mac DAW with an engineer built in. Say what you want and it makes the edit. Free download.

Get Audizer

Updated 2026-11-15. Corrections: support@intuzu.com