48 kHz or 44.1 kHz? Sample rate and bit depth for video audio

What sample rate and bit depth mean

Digital audio measures the sound wave many times a second. The sample rate is how many measurements per second; bit depth is how precisely each one is stored. MDN describes sample rate as the number of samples taken per second and resolution as how many bits encode each amplitude sample: 8 bits gives 256 levels. [2]

Sample rate sets the highest frequency you can capture. Audacity says the maximum reproducible frequency is about half the sample rate, the Nyquist frequency, and that 44.1 kHz covers frequencies up to about 20 kHz. [3] MDN adds that 48 kHz leaves a wider transition band (4 kHz versus 2.05 kHz at 44.1 kHz), which may preserve high frequencies slightly better. [2] For speech, both are more than enough.

Bit depth sets the noise floor and how much room you have before quiet detail is lost. MDN says most audio files use 16-bit integers, with 24-bit, 32-bit integer and 32-bit float also common. [2] Audacity processes audio internally in 32-bit float and notes that 32-bit files are twice the size of 16-bit. [5][6]

Why video standardises on 48 kHz

Video has historically used 48 kHz: MDN notes DVD audio is recorded at 48 kHz. [2] Today the strongest reason is the platform request. YouTube recommended upload settings ask for stereo or stereo plus 5.1 audio at 48 kHz. [1]

Recording tools follow. Riverside says its local recordings are uncompressed 48 kHz WAV. [7] Apple Podcasts accepts both 44.1 and 48 kHz, so 48 kHz works for a show that is published as video and audio. [8]

The exception is audio-first software. Audacity defaults to 44.1 kHz and calls that default highly recommended unless you have a reason to change it. [3] If you record voice-over in Audacity for a video, set the project to 48 kHz first.

What a mismatch actually does

A sample rate is a promise about how fast to play the samples back. If 48,000 samples recorded in one second are played as though they were 44,100 per second, the clip takes 48,000 / 44,100 = 1.088 seconds to play. Audacity documents exactly this effect: changing a track rate without resampling changes its speed and pitch. [4]

The numbers are not subtle. A 10-minute voice-over recorded at 48 kHz but read as 44.1 kHz plays for about 10 minutes 53 seconds (about 8.1% slower, so it runs 8.8% longer) and drops in pitch by about 1.47 semitones (12 x log2 of 48,000 / 44,100). The reverse mismatch makes it about 8.8% faster and higher, finishing in about 9 minutes 11 seconds.

Editors usually resample correctly on import, which avoids that, but resampling is extra processing. Audacity notes that tracks at other rates can coexist in a project and that playback and mixdowns all use the project rate. [5] The safest workflow is not to rely on conversion at all: record every source at the rate of the project.

If audio and picture slide apart gradually while the pitch sounds right, the sample rate is probably not the cause. Look at the video side next, starting with variable frame rate recordings, which our variable frame rate guide covers.

Bit depth: 16, 24 or 32-bit float

MDN says most audio files use 16-bit samples. [2] Apple Podcasts sets 44.1 kHz, 16-bit or 24-bit as the minimum for WAV and FLAC uploads, and recommends 24-bit. [8] For recording, 24-bit gives more room to set a conservative gain and still capture quiet passages cleanly.

For editing, 32-bit float avoids rounding errors during processing. Audacity uses 32-bit float throughout by default and says dither is never applied inside a 32-bit float project because no reduction in bit depth happens. [6] Dither is only added when you reduce bit depth, such as exporting to 16 or 24-bit. [6]

Storage is the trade-off. MDN gives the formula for uncompressed audio: channels x bytes per sample x sample rate. 48 kHz, 16-bit stereo is 192,000 bytes per second. [2] At 24-bit that becomes 288,000 bytes per second, about 1.04 GB per hour; at 32-bit float, 384,000 bytes per second, about 1.38 GB per hour.

Settings checklist: mic to export

Check each link in the chain once and write the settings down; a single 44.1 kHz device is enough to cause trouble.

• USB mic or interface: set 48 kHz in the system sound settings and the recording app.
• Recorder or DAW: set the project rate to 48 kHz before recording; Audacity defaults to 44.1 kHz. [3][5]
• Remote interviews: confirm the service records 48 kHz; Riverside states 48 kHz WAV. [7]
• Bit depth: record 24-bit where possible; edit in 32-bit float if available. [6][8]
• Editor: create the project at 48 kHz and check imported clips report 48 kHz.
• If a clip plays at the wrong pitch, resample it rather than changing its rate. [4]
• Export: 48 kHz audio for video uploads; YouTube asks for 48 kHz. [1]
• Podcast audio: 44.1 or 48 kHz both meet Apple Podcasts requirements. [8]

Sources & verification

Product details and prices can change. Check the linked provider before buying.

1. YouTube Help: YouTube recommended upload encoding settings Accessed 30 Oct 2026
2. MDN Web Docs: Digital audio concepts Accessed 30 Oct 2026
3. Audacity Manual: Sample Rates Accessed 30 Oct 2026
4. Audacity Manual: Audio Track Dropdown Menu Accessed 30 Oct 2026
5. Audacity Manual: Audio Settings Preferences Accessed 30 Oct 2026
6. Audacity Manual: Dither Accessed 30 Oct 2026
7. Riverside: FAQ Accessed 30 Oct 2026
8. Apple Podcasts for Creators: Audio requirements Accessed 30 Oct 2026