48000 Hz

48000 Hz (48 kHz) is a digital audio sample rate: the signal is measured 48,000 times per second, which lets it represent frequencies up to 24 kHz, above the upper limit of human hearing.

It is the standard rate for professional audio for video: film, television, streaming video, games and most broadcast systems. For music-only releases, its long-time rival is 44.1 kHz, the CD rate.

What does 48 kHz mean?

When an analogue signal is recorded digitally, an analogue-to-digital converter takes regular snapshots of its voltage. At 48 kHz it takes 48,000 snapshots (samples) every second, and on playback a digital-to-analogue converter uses those samples to rebuild the continuous waveform.

According to the sampling theorem behind all digital audio, a sample rate can represent frequencies up to half its value, a limit known as the Nyquist frequency. For 48 kHz that limit is 24 kHz. Human hearing tops out at around 20 kHz at best, so 48 kHz covers the whole audible range with room to spare.

In plain terms: a 48 kHz recording is a very fast series of measurements, fast enough to capture everything we can hear. It is not a measure of how “detailed” a sound is in the loudness sense; that is set by bit depth, such as 24-bit.

How it is defined

The Audio Engineering Society’s AES5 recommended practice on preferred sampling frequencies names 48 kHz as the rate for the origination, processing and interchange of professional audio programs using PCM. The current edition is AES5-2018, reaffirmed in 2023 (AES Standards News).

Names and meanings

You will see it written as 48000 Hz, 48,000 Hz, 48 kHz or simply “48k”. In a production context it always means the sample rate. Strictly speaking, a 48,000 Hz tone would be an ultrasonic sound far above hearing, but that is not what engineers mean when they talk about “working at 48k”.

Where 48 kHz came from

Before standards: many rates

In the late 1970s, early digital recorders used a range of proprietary sample rates, roughly from 32 kHz to 50 kHz, and standardisation work started in the early 1980s (48,000 Hz, Wikipedia). Two rates came out of that era and are still with us.

44.1 kHz: a by-product of video tape

The CD rate, 44.1 kHz, came from the way early digital audio was stored: on video cassettes, using PCM adaptors, because video recorders were the only affordable machines with enough bandwidth. Three samples per video line fitted both TV systems: 294 active lines × 50 fields × 3 = 44,100 for PAL, and 245 lines × 60 fields × 3 = 44,100 for NTSC (44,100 Hz, Wikipedia). The Compact Disc inherited it.

48 kHz: chosen for professional work and picture

Professional and broadcast engineers wanted a rate that worked cleanly with film and TV frame rates and with existing broadcast systems. 48,000 is evenly divisible by 24, 25 and 30 frames per second, and it sits in a simple 3:2 ratio with 32 kHz, a rate already used in broadcast transmission. It was adopted in the AES5 recommendation first published in 1984.

DAT and the professional studio

Digital Audio Tape (DAT), introduced in 1987, recorded at 48 kHz and reinforced it as the standard rate for professional audio (Digital Audio Tape, Wikipedia). Some early consumer DAT machines deliberately could not record at 44.1 kHz so they could not be used to clone CDs, one more reason why so many masters, jingles and broadcast recordings of that era were made at 48 kHz.

Today

Video has made 48 kHz the default for most of the audio world. It is the rate used for film and TV production and delivery, for broadcast delivery (which also follows loudness recommendations such as EBU R128), for Dolby Atmos masters, and it is what YouTube recommends for uploads (YouTube Help). Operating systems such as Windows and Android also commonly use 48 kHz as their default output rate.

How 48 kHz works

Room for the anti-aliasing filter

Before sampling, a converter must remove frequencies above the Nyquist limit; otherwise they fold back into the audible range as aliasing, a type of distortion that sounds unrelated to the music. This is the job of the anti-aliasing filter. At 44.1 kHz the filter has to go from passing 20 kHz to blocking 22.05 kHz, a transition band of just over 2 kHz. At 48 kHz the band is 4 kHz wide (20 to 24 kHz), which gives designers more room for a gentler filter. Modern oversampling converters handle both rates very well, so in practice the difference is small, but it is one of the engineering reasons behind the rate.

Whole samples per video frame

The main practical reason 48 kHz dominates video is simple arithmetic. Every frame of picture corresponds to a whole number of audio samples:

Frame rate	Use	Samples per frame at 48 kHz	At 44.1 kHz
24 fps	Cinema	2,000	1,837.5
25 fps	PAL / European TV	1,920	1,764
30 fps	Video, web	1,600	1,470
60 fps	Games, high frame rate video	800	735

At 24 fps, 44.1 kHz leaves half a sample per frame. Software can handle that, but whole numbers make editing, synchronisation and conversion between systems much simpler, which is why picture-based workflows standardised on 48 kHz.

Digital interfaces and clocking

Digital audio connections such as AES3 (AES/EBU), S/PDIF and ADAT carry 48 kHz audio, and every device in a system must run at the same rate, locked to a single word clock master. If an interface, a converter and a DAW disagree about the rate, you get clicks, dropouts or audio that plays at the wrong speed.

48.048 kHz: the film pull-up rate

Film and TV production has a special case. When material shot at true 24 fps is finished in NTSC-based video at 23.976 fps, picture is slowed by 0.1%. Recording sound at 48.048 kHz (0.1% faster) means that after the same pull-down the audio plays in sync at a true 48 kHz (Trew Audio). It is a specialist workflow that only works if the whole post-production team agrees on it before the shoot.

48 kHz vs 44.1 kHz

This is one of the most debated questions in home recording, and the honest answer is that the difference is about workflow, not sound quality.

	44.1 kHz	48 kHz
Origin	CD, via video-tape PCM adaptors	Professional audio, broadcast, DAT
Nyquist limit	22.05 kHz	24 kHz
Main home today	CD and many music-only workflows	Video, film, TV, games, streaming video, Atmos
Video frame maths	Fractional at 24 fps	Whole samples at 24, 25, 30 and 60 fps
Data rate	Baseline	About 9% more samples per second

Both rates capture the full range of human hearing. Most individual listening tests have found that listeners cannot reliably hear CD-quality conversion when it is inserted into a high-resolution chain. In the best-known study, Meyer and Moran ran 554 trials in which listeners had to detect a 16-bit/44.1 kHz stage added to high-resolution playback, and got 276 right (49.8%), no better than chance (Journal of the AES, September 2007). A 2016 meta-analysis did find a small effect among trained listeners (see 96000 Hz), but either way, sound quality is not a reason to choose 48 kHz over 44.1 kHz or the reverse.

Which one should you use?

• Anything for picture (film, TV, ads, YouTube video, games): 48 kHz, no question.
• Music that may end up in video (music videos, sync licensing, social media, Dolby Atmos mixes): 48 kHz is the safer default today.
• Music-only projects built from 44.1 kHz samples and aimed at streaming and CD: 44.1 kHz is perfectly fine and avoids converting your sample library.
• Heavy sound design with extreme pitch-shifting, or when you want lower buffer latency: consider higher rates such as 96 kHz, then deliver at 48 kHz.

Whatever you choose, choose it at the start and stay consistent. The rate matters far less than avoiding unnecessary conversions in the middle of a project.

48 kHz in practice

Delivery specifications

Sample rate is only part of a delivery spec; loudness and true-peak limits usually matter just as much. Some common targets:

Destination	Sample rate	Loudness and peaks
European broadcast (loudness per EBU R128)	48 kHz	−23 LUFS integrated, maximum −1 dBTP (EBU R128)
US broadcast (loudness per ATSC A/85)	48 kHz	−24 LKFS integrated
Netflix	48 kHz, 24-bit in film/TV workflows	−27 LKFS dialogue-gated (±2 LU), peaks no higher than −2 dBTP (Production Expert)
YouTube uploads	48 kHz recommended	Playback is loudness-normalised by the platform
Music streaming (distribution)	Native project rate, commonly 44.1 or 48 kHz at 24-bit	Platform normalisation; leave true-peak headroom for lossy encoding

For music, Apple’s Apple Digital Masters guidance asks for high-resolution masters at their original rate rather than files that have been upsampled (Apple Digital Masters). The same principle applies everywhere: deliver at the rate you worked at, and convert only once, at the end, if a destination requires it.

Across genres and media

• Film, TV, games and broadcast: 48 kHz throughout, sometimes with source recording at 96 kHz for sound design before delivery at 48 kHz.
• Pop, hip-hop, trap and EDM: many producers still work at 44.1 kHz because their sample libraries are at 44.1 kHz, but sessions intended for video, Atmos or sync are increasingly started at 48 kHz.
• Classical and jazz: often recorded at higher rates for archiving, then delivered at 48 kHz or 44.1 kHz for distribution.

Setting 48 kHz in your DAW

Set the rate before you record or import anything. Most DAWs convert imported files automatically, but it is better to know what is happening than to let a mismatch slip through.

DAW	Where to set it
Pro Tools	Chosen when creating the session; to change it later, save a copy of the session at the new rate
Logic Pro	File > Project Settings > Audio > Sample Rate
Ableton Live	Settings (Preferences) > Audio > In/Out Sample Rate
FL Studio	Options > Audio settings > Sample rate
Cubase / Nuendo	Project > Project Setup > Sample Rate
Fender Studio Pro (formerly Studio One)	Song > Song Setup
Reaper	Project settings, and the sample rate in the audio device preferences

Also check your audio interface control panel and operating system. If another application or the system audio settings hold the interface at a different rate, your DAW may refuse to switch or may resample in the background.

Tools that matter at 48 kHz

• Field recorders such as those from Sound Devices and Zoom default to 48 kHz for film and broadcast work, often with 32-bit float recording for safety.
• Sample rate converters: dedicated tools such as iZotope RX and Weiss Saracon, and the high-quality converters built into modern DAWs, handle conversion between 48 and 44.1 kHz when a delivery requires it.
• True-peak limiters and meters: oversampled meters, as defined in ITU-R BS.1770, show the peaks that appear between samples after conversion or lossy encoding.

Common mistakes and myths

Mistakes

• Mismatched rates in one session: importing 44.1 kHz audio into a 48 kHz project without conversion makes it play noticeably fast and sharp (by about 1.5 semitones), and it drifts out of sync with picture.
• Changing the rate mid-project: every conversion is an extra processing step and a chance for mistakes. Decide at the start.
• Converting at 0 dBFS: converting a master that already peaks at 0 dBFS can create inter-sample peaks above full scale. Leave some true-peak headroom (for example −1 dBTP) before conversion and lossy encoding.
• Forgetting the interface: a DAW set to 48 kHz while the interface or system runs at 44.1 kHz causes clicks, errors or silent resampling.

Myths

• “Digital audio is a staircase.” The samples are points, not steps. After the reconstruction filter, the output is a smooth continuous waveform; a correctly converted 48 kHz signal reproduces everything below 24 kHz.
• “Upsampling improves quality.” Converting a 44.1 kHz file to 48 kHz or 96 kHz cannot add detail that was never recorded. At best it changes nothing, at worst a poor converter adds artefacts.
• “48 kHz sounds better than 44.1 kHz” (or the reverse). Both cover the audible range. Choose based on where the audio is going.

Standards and research

Standard or study	What it says about 48 kHz
AES5 (first edition 1984; AES5-2018, reaffirmed 2023)	Recommends 48 kHz as the rate for professional PCM audio.
EBU R128 (2010) and EBU Tech 3341	European broadcast loudness standard; programmes normalised to −23 LUFS with a −1 dBTP maximum.
ITU-R BS.1770 (first published 2006, current revision BS.1770-5)	Defines loudness and true-peak measurement used by R128 and A/85.
ATSC A/85	US TV loudness practice supporting the CALM Act; −24 LKFS target.
Meyer and Moran, JAES 2007	Listeners could not detect a 16-bit/44.1 kHz stage inserted into high-resolution playback (554 trials, 49.8% correct).

Related terms

• Sample rate: the general concept behind 48 kHz.
• 44100 Hz: the CD rate and 48 kHz’s closest alternative.
• 192000 Hz: the highest common studio rate.
• Nyquist frequency: why 48 kHz can represent up to 24 kHz.
• Anti-aliasing filter: the filter that makes sampling work.
• 24-bit: the bit depth usually paired with 48 kHz in professional delivery.
• Oversampling: how modern converters and plugins work at higher internal rates.

FAQ