what it estimates
What does an audio latency calculator measure?
An audio latency calculator turns a digital buffer into a time value. A buffer is a block of samples that the computer waits to collect or process. At a fixed sample rate, a larger block takes longer to pass through the audio path, so it generally creates more delay but gives the system more time to avoid glitches.
This is a planning estimate, not a promise that every device will report the same number. Drivers, converter safety buffers, Bluetooth codecs, plugins, displays, and browser scheduling can add delay that is not visible in the basic buffer setting. Use the result to choose what to test next, then compare the real device with the Audio Latency Test or a dedicated loopback measurement.
One bufferUseful for a one-way playback or monitoring estimate when you want to understand the time of a single audio block.
Two buffersA practical starting point for an input-to-output round-trip estimate, before converter and driver overhead are added.
Extra delayUse this field for a known plugin, codec, capture device, or other delay that is outside the buffer calculation.
the formula
Audio latency formula: buffer size and sample rate
The basic formula is: buffer samples ÷ sample rate × 1000 = milliseconds for one buffer. Multiply that value by the number of buffer passes, then add any known extra delay. For example, a 256-sample buffer at 48,000 Hz is about 5.33 ms for one buffer and about 10.67 ms for two passes before extra processing is included.
The calculator keeps the arithmetic visible so you can audit the result. If your interface or DAW reports a round-trip value that is higher, that difference can come from input and output buffers, converter latency, driver safety margins, USB or network transport, or plugin delay compensation.
| Buffer | 44.1 kHz, one pass | 48 kHz, one pass | 48 kHz, two passes |
| 64 samples | 1.45 ms | 1.33 ms | 2.67 ms |
| 128 samples | 2.90 ms | 2.67 ms | 5.33 ms |
| 256 samples | 5.80 ms | 5.33 ms | 10.67 ms |
| 512 samples | 11.61 ms | 10.67 ms | 21.33 ms |
read the number
One-way latency versus round-trip latency
One-way latency describes a signal moving from a playback source to an output. It can help explain why a larger buffer feels less immediate during playback or software monitoring. Round-trip latency describes an input being captured, processed, and returned to an output. It is the more useful comparison for a microphone, guitar, vocal monitor, or interface workflow.
Do not treat the number as a universal quality score. A stable 25 ms path may be workable for a call, while the same delay can feel distracting for live vocals or rhythm-sensitive performance. A lower buffer can reduce delay but may increase CPU pressure, clicks, dropouts, or unstable audio.
RecordingStart with the round-trip option, then confirm the real result with the Mic Latency Test and your interface software.
HeadphonesCompare wired and wireless paths with the Headphone Latency Test; a Bluetooth codec is not represented by buffer size alone.
Video syncFor a TV, capture card, or soundbar, use the estimate as a starting point and then check the actual offset with the Audio Sync Test.
use it in a workflow
How to use the result when fixing audio delay
First record the current buffer, sample rate, output device, connection type, and the calculated estimate. Change one setting at a time. If you reduce a buffer from 512 to 256 samples, the theoretical time for each buffer is cut in half, but the real path may still be late because of a wireless codec, a plugin, a display, or a capture device.
Next, run a repeatable real-world check. Use the same browser, device, volume, and content before and after the change. The site's audio delay testing guide explains how to isolate a fixed delay from drift; the audio interface latency guide covers drivers, buffer choices, direct monitoring, and plugin overhead.
Change one variableKeep the sample rate and output device fixed while comparing buffer sizes.
Watch stabilityA smaller theoretical number is not a successful fix if it causes clicks, dropouts, or jitter.
Retest the real pathUse a practical browser test or loopback measurement after the calculation, especially for live monitoring.
accuracy and privacy
Calculator limits and local processing
This calculator does not inspect your driver, measure your hardware, or claim laboratory accuracy. It performs a transparent formula in your browser. The values stay in the page and are not uploaded as part of the calculation. If you use the microphone test or another browser permission separately, that permission belongs to that test rather than this calculator.
Use a dedicated loopback tool or the values reported by your interface when you need a certified studio measurement. For ordinary troubleshooting, the estimate is useful because it gives you a baseline: you can see whether a buffer change should make a meaningful difference before listening to the same setup again.
Not a lab measurementConverter, driver, transport, codec, display, and plugin delay may sit outside the formula.
Keep a baselineWrite down the settings and repeat the same test after each meaningful change.
Prefer stable audioA slightly higher delay can be better than a lower setting that crackles or drops samples.
continue testing
Use the estimate with an audio latency test
quick answers
Audio Latency Calculator FAQ
What is the formula for audio latency?
For one buffer, divide buffer samples by sample rate and multiply by 1000. Multiply by the number of buffer passes and add known extra delay for a simple path estimate.
How many milliseconds is a 256-sample buffer at 48 kHz?
One 256-sample buffer at 48 kHz is about 5.33 ms. A simple two-buffer round-trip estimate is about 10.67 ms before converter, driver, plugin, codec, or device overhead.
Does the calculator measure my actual interface latency?
No. It calculates a theoretical estimate from the values you enter. Use a real loopback or microphone test to check the complete hardware path.
Should I use one buffer or two buffers?
Use one for a one-way playback or block-time estimate. Use two as a starting point for input-to-output round-trip latency, then verify it with the device or interface you actually use.
Why is my measured latency higher than the calculator result?
The real path can include converter safety buffers, drivers, transport, plugins, Bluetooth codecs, displays, capture devices, and browser scheduling that are not included in the basic buffer formula.