Why does 44.1 kHz capture up to 22.05 kHz?

Because the limit is exactly half the rate. The Nyquist theorem requires more than two samples per cycle to represent a frequency, so the top frequency a rate can carry is half of it, and 44.1 kHz therefore reaches 22.05 kHz. That covers the roughly twenty hertz to twenty kilohertz span of human hearing.

Is 96 kHz better than 48 kHz?

Not in any way you can hear in a finished file. Higher rates give a converter more room to place its anti-aliasing filter, which matters during processing and recording, but they do not add audible content beyond human hearing. Delivering at 96 kHz mostly means larger files for the same result.

Does this tool resample my audio?

No. It is a static table and nothing is loaded. Changing a file's rate is a conversion job with its own interpolation and filtering step, and the reference exists only to help you choose the destination rate before you run one.

What does the captures-up-to column really mean?

It is the theoretical Nyquist frequency, which is simply each rate halved. A real converter places a lowpass anti-aliasing filter a little below that number so content above it cannot fold back down into the audible band, so genuine content never extends right up to the quoted limit.

Which rate should I choose for a podcast?

Either 44.1 kHz or 48 kHz is more than enough, because speech carries very little energy above about ten kilohertz. Matching the rate of whatever you record and edit at is the useful rule; a lower rate would only save space and would cost you nothing audible, but it also buys you almost nothing in a compressed container.

Sample Rate Reference

You Might Also Need

Sample Rate Calculator

Calculate sample rate requirements

Audio Format Converter

Universal audio format converter with batch support

Bit Depth Calculator

Calculate bit depth needs

WAV Converter

Convert audio to lossless WAV format

Sample Rate Reference explains what the common sample rates buy you, from 8 kHz telephony up to 192 kHz archive work, with the frequency range each one can represent written beside it. It is a static reference: no audio is loaded, measured or converted.

What is Sample Rate Reference?

Nine rates are listed, each with its capture limit and the place it is normally used. A row of filter buttons switches the view, and nothing on the page reads or alters a file.

What the filters show:

All shows every rate in the table.
Speech keeps rates of 32 kHz and below.
Music keeps 44.1 kHz, 48 kHz and the high-resolution entries.
Video keeps 48 kHz and 96 kHz.
Hi-res keeps everything from 88.2 kHz upward.

The rows themselves read as a history of audio delivery:

8 kHz and 16 kHz, the telephony and voice-over-IP rates that carry speech and nothing above 4 kHz or 8 kHz.
22.05 kHz and 32 kHz, legacy low bitrate audio and broadcast contribution links.
44.1 kHz, the CD rate and the default for music delivery, capturing up to 22.05 kHz.
48 kHz, the video and film production standard, capturing up to 24 kHz.
88.2 kHz and 96 kHz, high-resolution production rates that relax converter filtering.
192 kHz, a specialist measurement and mastering rate with no audible benefit on delivery.

Two rows are worth committing to memory. 44.1 kHz is the number behind the CD and most streaming music, and 48 kHz is the number behind television, film and most professional recording interfaces; choosing the wrong one of the two is the usual reason a file gets resampled for no benefit at all. The second column of the table is simply each rate halved, and that halving is the whole theory: the Nyquist theorem says a frequency can only be represented when it is sampled more than twice per cycle, so the highest frequency a rate can carry is half the rate itself. That is why 44.1 kHz is quoted as covering the roughly twenty hertz to twenty kilohertz span of human hearing. In practice a converter places a lowpass anti-aliasing filter just below that limit so content above it cannot fold back down into the audible band, and higher rates exist largely to give that filter more room rather than to add content anyone can hear. Resampling between two rates is a separate operation with its own interpolation and filtering, and none of it happens here.

Where the reference stops being useful:

The capture limit shown is the theoretical Nyquist frequency, and a real converter's anti-aliasing filter always sits a little below it.
Higher rates do not add audible content, so 96 kHz of a poor recording is still a poor recording.
The five filter buttons are fixed categories over the table's own rows rather than an analysis of anything.
Nothing is loaded, so the page cannot read the sample rate of a file you have in front of you.
Changing a file's rate is a conversion job with its own interpolation step, which is not what this page does.
The list is a conventional set; rates such as 176.4 kHz and pull-down values like 47.952 kHz are deliberately absent.

For almost every job only two numbers matter. Use 44.1 kHz when the deliverable is music and 48 kHz when it is video or a production session, because matching the project rate avoids a resample entirely. Reach for the low rows when the material is speech and the file has to stay small, and treat the high-resolution rows as a session decision rather than a delivery one: they exist to keep processing and filtering comfortable, not to make the finished file sound better on a phone.

How to use Sample Rate Reference

Pick the filter that matches the job at hand: Speech, Music, Video or Hi-res, or leave it on All.
Read the capture column to see the highest frequency the rate can represent before the converter filters anything.
Match the rate to the destination project, using 44.1 kHz for music and 48 kHz for video work.
Note the rate you settled on and take it into a converter, since no file is changed on this page.

When to use Sample Rate Reference vs related tools

Sample Rate Reference settles the rate before a session starts, which is when the decision is cheapest to make. Sample Rate Calculator works the other way round, deriving requirements from a frequency range or a duration, while Audio Format Converter and WAV Converter are where a file actually gets resampled into the setting you chose. When the question is really about how many bits each sample carries rather than how often samples are taken, Bit Depth Calculator covers that axis on its own.

Privacy & Security

This tool runs entirely in your browser — no data ever leaves your device. There is no server round-trip, no upload, no logging, and no account required. Your input is processed locally using client-side JavaScript and is never stored, transmitted, or accessible to anyone else. When you close the tab, everything disappears.