What Is Sample Rate in Audio? 44.1 kHz vs. 48 kHz Explained

The sample rate is the number of times per second a sound is measured when it is recorded digitally. It sets the highest frequency that can be stored: half the rate. 44.1 kHz is the standard for music and 48 kHz the standard for video, and both cover everything we can hear. Use the rate your destination expects, and convert a file properly when it has the other one.

What Is Sample Rate?

Sound in the air changes continuously. A computer cannot store something continuous, so a recording device measures the sound wave at regular intervals and stores each measurement as a number. One measurement is a sample, and the number of samples per second is the sample rate. At 44.1 kHz there are 44,100 samples every second for each channel.

It helps to think of film, which shows 24 still pictures per second and looks like movement. Audio needs far more steps per second than pictures do, because sound waves change much faster than anything we see. The other number that describes a recording, the bit depth, says how precisely each sample is stored; it affects the noise floor, not the frequency range.

Why 44.1 kHz? The Sampling Theorem

How many samples are enough? The sampling theorem gives the answer: to capture a frequency, you need more than two samples for each of its cycles. A sample rate can therefore store frequencies up to half its own value, a limit called the Nyquist frequency. Human hearing ends at about 20 kHz, so a rate above 40 kHz covers all of it.

The exact figure of 44.1 kHz is a piece of history. Early digital recordings were stored on video tape, and 44,100 samples per second was what fitted the picture lines of both television systems of the time. When the CD was specified, it kept that rate, as the history of the 44.1 kHz standard records. Its Nyquist frequency of 22.05 kHz leaves a little room above 20 kHz for the filter that removes everything higher.

44.1 kHz vs. 48 kHz: Comparison

Why Video Uses 48 kHz

Professional audio and video settled on 48 kHz for practical reasons. One of them is arithmetic: 48,000 divides evenly by the common frame rates, giving 2,000 samples per frame at 24 frames per second, 1,920 at 25 and 1,600 at 30, while 44,100 divided by 24 is 1,837.5. Cameras, video editors and broadcasters therefore expect 48 kHz, and it is the safe choice for anything that ends up next to a picture. Music distributors and streaming services, on the other hand, work with 44.1 kHz masters as a matter of course.

What Happens with the Wrong Sample Rate

Most software converts quietly when a file does not match the project, and you notice nothing. Problems appear when a file is read at the wrong rate without conversion. A 44.1 kHz recording played as if it were 48 kHz runs about 9 percent too fast and sounds roughly one and a half semitones higher; the reverse is slow and low. If speech in your video sounds slightly squeaky or sluggish, this is the first thing to check.

A different problem is drift, in which sound and picture start together and slowly move apart. It comes from two devices recording on their own clocks, and converting the sample rate does not cure it; the audio has to be stretched or re-synchronized in the editor.

How to Convert the Sample Rate

When a file has to match a project or a platform, convert it once, from the original.

Converting up does not improve a recording, and converting back and forth several times gains nothing, so keep the original file and convert from it whenever you need another rate.

Do Higher Sample Rates Sound Better?

Rates of 88.2, 96 and 192 kHz can store frequencies far beyond the limit of hearing. That is useful in production, for example when a recording is slowed down a lot or processed heavily, and some engineers record at 96 kHz for that reason. For listening, any benefit is small and disputed, and the files are two to four times as large. For finished music and video, 44.1 and 48 kHz are enough.