44.1 or 48 kHz? Choosing an Audio Sample Rate for Video

Open the export settings of almost any editor and you will find a menu offering 44.1 kHz and 48 kHz, usually with no explanation. Most people leave the default, and most of the time that is fine. The reason to understand it is the occasions when it is not: a clip that drifts out of sync, a pitch that sounds slightly off, or an export that is quietly resampled in a way that costs a little quality.

The two numbers are not a quality choice, in the way that bitrate is. They are a convention, and the real rule is simple: keep everything in one project at one rate, and prefer 48 kHz for video.

What a sample rate is

Digital audio is a series of measurements of a sound wave, taken many times a second. The sample rate is how many. 44.1 kHz means 44,100 measurements every second, and 48 kHz means 48,000. A higher rate can in principle capture higher frequencies, but both are well above what human hearing needs, which tops out around twenty kilohertz. The difference between them is inaudible to nearly everyone in ordinary listening.

Where each one comes from

The video convention is the one that matters here. 48 kHz is the standard for video because it fits neatly with the frame rates used in film and television, which keeps picture and sound aligned over long durations. When you export a video, 48 kHz is almost always the right setting.

What goes wrong when they are mixed

If a project is set to one rate and you import audio at another, the editor must convert it, and a good editor does so automatically and invisibly. A poor conversion, or a player that ignores the setting, can cause two kinds of fault. The audio can play at the wrong speed, which shows up as a change of pitch and a slow drift against the picture. Or it can be resampled at a quality cost that is small but real.

• A pitch that is slightly too high or too low suggests a sample rate mismatch.
• Audio that slowly slides out of sync over a long clip can have the same cause.
• A thin, slightly dull sound can come from a poor quality conversion.
• A problem that appears in one player and not another points at the player.

A practical rule

1. Set the project to 48 kHz if the output is video.
2. Record or generate your audio at 48 kHz where you can.
3. If a source is 44.1 kHz, let the editor convert it once, on import, not repeatedly.
4. Export at 48 kHz and leave it there.

Does a higher rate sound better?

Not for delivery. 96 kHz has real uses in recording and production, where it gives processing more room, but a finished video played back through ordinary speakers gains nothing from it, and the file is larger. For a video that will be streamed or shared, 48 kHz is more than sufficient, and nobody can tell it apart from a higher rate in a listening test.

Checking what a file uses

The video file info tool on this site reports the audio sample rate of a file, alongside its codec and channel count. If you are chasing a sync or pitch problem, it is the first thing to check: a file whose audio is 44.1 kHz inside a project that assumed 48 kHz explains a great deal in a single look.

If you generate narration with the voiceover tool, the output is already at a standard rate for web use. The thing to watch for is only the moment it meets your editor, where the project setting decides how it is treated.