Audio latency is measured in milliseconds (ms). Under 20 ms usually feels immediate, 20–50 ms is workable for many tasks, 50–100 ms becomes noticeable, and delay above 100 ms often disrupts gaming, monitoring, calls or lip sync.
Usually feels immediate for playing, monitoring and competitive use.
Often acceptable for casual gaming, calls and general listening.
Timing starts to feel soft; speech and video may look slightly detached.
Echo, late feedback and obvious lip-sync errors are common.
What audio latency actually means
Audio latency is the total delay created while sound is captured, processed, transmitted and played. A tap, spoken word or on-screen event happens first; the audible result arrives later.
One-way output latency matters for video and gaming. Round-trip latency includes input and output, so it matters most when you hear your own microphone or instrument through software. Use the mic latency test when monitoring is the problem.
Time from a device or app sending audio until your headphones or speakers reproduce it.
Time from a microphone or instrument entering the system until software receives it.
Input delay plus processing and output delay—the number performers usually feel.
Difference between the visible event and the sound; test it with the audio sync tool.
How much audio latency is acceptable?
There is no single perfect number. The acceptable limit depends on whether you are listening passively or reacting in real time. Music playback can tolerate buffering; live monitoring cannot.
For spoken calls, consistency often matters as much as the raw number. A stable 70 ms delay can feel easier than delay that jumps between 30 and 120 ms.
Aim below 20 ms round trip; many performers prefer roughly 10 ms or less.
Lower is better; below 30–40 ms audio output helps cues feel connected.
Lip sync is usually comfortable when offset stays within roughly 40–60 ms.
Small local device delay is fine, but network delay adds to the conversation gap.
What causes audio delay?
Latency accumulates across the entire signal chain. The biggest contributor can be the wireless codec, an oversized software buffer, effects processing, operating-system mixing, a television's picture processing, or the network.
Bluetooth delay deserves its own diagnosis because codec, earbuds and device behavior interact. Measure it on the Bluetooth latency test page before changing settings.
Bluetooth encoding, radio transmission and decoding add buffering.
Larger buffers prevent glitches but increase delay.
Noise removal, spatial audio, EQ, plugins and resampling need time.
TV motion smoothing and video pipelines can move pictures later than sound.
How to reduce audio latency
Change one part of the chain at a time, then retest. This makes it clear which adjustment actually helps and prevents trading stable audio for clicks or dropouts.
- Measure a baseline
Run the main audio latency test with the same device, browser and connection you normally use.
- Use a wired path
For recording or rhythm-sensitive work, wired headphones and a direct audio interface are the most reliable shortcut.
- Lower the buffer carefully
Reduce buffer size in the DAW or audio app until delay improves without crackles.
- Disable unnecessary processing
Temporarily turn off spatial audio, enhancement suites, noise suppression and heavy plugins.
- Match the media path
Use game/low-latency mode on earbuds or TVs and avoid routing audio through extra devices.
- Retest and record the result
Compare the new measurement with the baseline instead of judging from memory.
How to test audio latency accurately
Browser tests provide a practical estimate, not laboratory certification. Keep the environment consistent, repeat the test several times and use the median rather than one unusually fast or slow result.
Choose the test that matches the symptom: the main tap test for perceived output delay, the headphone test for wired or wireless listening, the microphone round-trip test for monitoring, and the A/V sync test for lip-sync errors.
Use at least five attempts and ignore accidental taps or obvious outliers.
Do not switch codec, output device, browser tab or enhancement settings between runs.
Measure the TV, console, earbuds or interface exactly as you normally connect it.
Audio hardware latency and internet latency are different problems.
Measure your current setup
Use the browser tool to establish a repeatable baseline before changing devices or settings.
Audio latency FAQ
Is 100 ms audio latency bad?
For passive listening it may be tolerable, but 100 ms is clearly noticeable for gaming, live monitoring, calls and lip sync.
Can Bluetooth audio have zero latency?
No practical Bluetooth path is truly zero-latency. Low-latency modes and newer codecs can reduce delay, but the phone, codec and earbuds all matter.
Why do I hear my voice late in headphones?
You are hearing round-trip latency: microphone input, software processing and headphone output delay combined.
Does a faster internet connection reduce audio latency?
It can reduce network delay in calls or cloud gaming, but it does not fix local Bluetooth, buffer or device-processing latency.
What is the difference between latency and lag?
They are often used interchangeably. Latency is the measurable delay; lag is the user-visible or audible effect of that delay.