Audio and video out of sync on a 24/7 prerecorded YouTube stream can come from a steady timing offset, drift that worsens over time, or a delay introduced in playback or delivery. Work out which pattern you have and where it appears before changing an OBS sync setting.
A fixed offset may call for a measured timing adjustment; accumulating drift calls for checking the file, playback path and sample-rate consistency. If a local encoded test is in sync but YouTube playback seems late or unstable, investigate stream health and delivery rather than shifting the audio source.
Find where the sync problem appears
Start by describing what you see and hear, not by moving a slider. Choose an event that is easy to match, such as a spoken word with a visible mouth movement, a bell strike, a drum hit, or a door closing. Note whether the sound comes before or after the image. A vague impression that the stream “feels late” is not enough to distinguish an audio offset from ordinary delay between the encoder and a viewer.
Check at least two outputs if your setup permits it: the local media or encoder output, and the actual YouTube playback or stream monitor. OBS preview and local monitoring are useful, but they do not necessarily show the timing of the encoded stream that viewers receive. YouTube’s guidance recommends testing with representative movement and sound, and checking the live stream rather than relying only on a preview. See YouTube’s encoder testing and stream-health guidance.
Also establish what is being played. Is it one video file, a playlist, a VLC source, or a sequence controlled by another application? Confirm the expected audio track is selected and that sound is not also coming from a second source. If you are using a playlist, the VLC playlist repeat setup is a useful reference for understanding how the source behaves at item changes and loop boundaries.
A problem heard only on YouTube is not automatically a source-sync problem. YouTube latency is the time between capture and a viewer receiving the stream; it is not a measure of the relative timing of audio and video. A stream can have a long player delay and still have sound and image aligned. Conversely, audio and video can be misaligned even when the player delay is short.
Compare an event near the start with one later on
Use the same kind of clear event near the beginning of the programme and substantially later. If the programme loops, check an event after a loop boundary as well. You do not need specialist equipment: write down whether the sound leads or trails, and estimate the gap consistently by counting frames or using the player timeline where practical. Avoid changing settings between observations.
The pattern is the key. If the gap remains roughly the same, the error is consistent with a fixed offset. If the gap grows as the programme continues, that points towards an ongoing timing problem rather than a single constant delay. A source can also appear correct before a loop and wrong afterwards if the playlist changes tracks, restarts playback, or switches to a file with different timing. These observations guide diagnosis; they are not a universal guarantee about the cause.
| What you observe | First line of investigation | Avoid doing first |
|---|---|---|
| Similar gap near the start and later | Measure which side leads and check the responsible source timing | Applying an arbitrary offset by trial and error |
| Gap grows over time | File timing, playback behaviour, sample-rate path and other ongoing processing | Trying to hide the start-of-stream gap with a constant delay |
| Local encoded output is aligned, YouTube playback seems delayed | Stream health, network and player delivery conditions | Shifting the source audio just because the viewer sees it later |
| Timing changes at a playlist or loop boundary | Track selection, source restart and file-to-file consistency | Assuming one setting will correct every item |
Keep a short record of the event, elapsed programme time, output checked and whether sound leads or trails. This makes it easier to compare before and after a change, especially when you are testing a continuous channel outside its normal viewing hours.
Measure a fixed audio offset
When the gap appears stable, measure it as carefully as you can before adjusting anything. Pick an event with a sharp sound and a visible moment that corresponds to it. A soft musical onset or a scene cut can be hard to judge; a hand clap, drum strike or clear consonant is usually easier. Repeat the check rather than relying on a single impression.
Record both direction and approximate size. For example: “the sound arrives after the visible hand clap by about the same amount near the beginning and later.” Do not write only “audio is late”: the correction depends on which part leads and on how the encoder interprets its timing controls. If you cannot tell which side is early, compare against the original file played locally before changing the stream configuration.
Check the file itself at the beginning and near the end. A prerecorded file may already contain a sync problem, or the file used by OBS may not be the version you thought you selected. Verify the filename, the audio track and the playback speed. For a playlist, compare more than one item; a correction that suits one file may make another worse.
If the file is in sync locally but the encoded output is not, the source routing or encoder path becomes more relevant. If the same mismatch is present in the original file, correct or replace the media before compensating in OBS. A source timing adjustment can make one output appear right while leaving the underlying media issue in place.
Adjust the relevant OBS source timing
OBS supports audio sync offset controls for sources, but the right source, direction and amount depend on the measured mismatch and on your routing. Change the source responsible for the sound rather than applying a global adjustment without knowing what else it affects. If audio is embedded in a media source, inspect that source; if sound is routed separately, trace that route before changing an offset. OBS documents its source controls in the source reference.
Make one measured change, then repeat the same event comparison in the encoded output. Do not copy a number from another channel: different files, source paths, encoder arrangements and audio routing can produce different results. A fixed offset is a timing correction, not a diagnosis, and it cannot correct a gap that keeps increasing during playback.
For a channel that must stay live, avoid experimenting casually on the production broadcast. There is no universal no-interruption procedure for changing media and audio settings while every possible 24/7 setup is running. When possible, reproduce the issue in a controlled test or test stream, note the existing configuration, make one change at a time and verify the result before applying it to the continuous channel. If you need to schedule a test or interruption, explain the plan to viewers where appropriate.
If the source is a playlist or looping setup, test the correction through the whole sequence and across a restart. Source activation and restart behaviour can change when a file begins playing again. The guide to looping a YouTube live playlist can help you examine the mechanics of repeated playback, but still test your own source configuration.
Investigate growing drift and sample-rate consistency
A gap that widens is different from a stable offset. Adding a constant delay might make one moment line up, but it will not keep later moments aligned if the timing continues to diverge. First test whether the file itself stays in sync when played from beginning to end outside the stream. Check the selected audio track, playback speed and any playlist transitions. Confirm that the source is not being restarted or routed through an unexpected duplicate audio path.
Then check the audio format settings through the path from the source to OBS. A sample-rate mismatch between an audio device and OBS is one plausible cause to investigate, though it is not the explanation for every drifting stream. OBS community support has discussed inconsistent sample rates as a possible drift factor; treat that as a troubleshooting lead, not an official universal diagnosis. YouTube’s encoder guide lists 44.1 kHz for stereo and 48 kHz for 5.1 audio in its recommendations. Use the setting appropriate to your output and keep the source, device and OBS configuration coherent.
If a device is not needed for prerecorded playback, check whether its audio is being captured unintentionally. A microphone, desktop audio path or monitoring loop can add a second signal that sounds like an echo or timing fault. Mute or remove one path only in a controlled test, then check the encoded result. Keep a note of what changed so you can restore the working configuration if the test makes matters worse.
Long-running tests matter because a brief preview may not reveal a small accumulating error. Compare the same sort of event after enough playback to make the trend apparent, including after a loop if the channel repeats content. If drift appears only after a restart or at a boundary, focus on source and playlist behaviour before changing sample-rate settings. A fixed offset is not a substitute for identifying an ongoing timing change.
Check local encoded output before YouTube playback
Before blaming the YouTube player, determine whether audio and video are already misaligned in the encoder’s output. A local recording of a short test can help, provided it captures the encoded output and not merely the source preview. You can also use a controlled test stream and inspect YouTube playback. Include representative motion and sound; the YouTube guidance specifically recommends testing content similar to what you will broadcast.
If the local encoded recording is aligned but YouTube playback appears delayed, do not treat the whole-stream delay as relative desync. Compare the audio and video within the playback itself: does a visible clap still align with its sound, even though the viewer receives it later than the encoder produced it? If yes, the issue is more likely about latency or delivery than source timing. YouTube explains that lower latency means less player read-ahead buffering and may make viewers more likely to notice interruptions; network congestion can also delay a live stream. Read YouTube’s explanation of latency and buffering.
If a YouTube recording or monitor shows the same relative mismatch as the local encoded output, return to the file and source path. If only one playback device shows a problem, compare another device or browser before changing a live source. The point is to find the first stage where alignment changes, not to assume that the viewer’s delay identifies the component at fault.
Inspect stream health and delivery conditions
Once you know the encoded audio and video are aligned locally, review the stream-health information and the delivery path. OBS describes dropped frames as a sign that the connection is unstable or cannot keep up with the configured bitrate. YouTube health messages can also identify ingest or stream configuration issues. These conditions may affect continuity or viewer experience; they do not by themselves prove that the audio source is out of sync.
Look for messages around the time viewers report trouble, and compare them with your own observations. Note whether the stream stalls, buffers or loses frames, and whether the relative sound-to-image gap actually changes. If the link is unreliable or the configured bitrate is too demanding for the available connection, address those delivery conditions separately. The bitrate guidance for 24/7 streaming on Indian broadband can help you review that side of the setup; it does not replace a sync test.
YouTube’s live stream health message reference describes categories of stream-health messages. Use the current guidance rather than assuming an old screenshot or a viewer comment identifies the fault. A comment such as “the sound is late” is a useful report to investigate, but ask whether the gap grows, whether it occurs on another device, and whether the same event is aligned in a local encoded test.
Verify the correction without guessing
After a change, repeat the original comparison: same event type near the beginning, later in the programme, and after a loop boundary if applicable. Check the output that matters to viewers, not only the OBS preview. If you corrected a steady gap, confirm that the direction is right and the gap remains stable. If you investigated drift, run a sufficiently long test to see whether the gap still grows.
Change one thing at a time. If you alter an offset, sample rate and playlist behaviour together, a better result will not tell you which change mattered, and a worse result will be harder to undo. Keep the before-and-after notes and the settings you touched. This is particularly useful on a devotional or study channel where an unnoticed overnight change can affect hours of playback.
For a long-running channel, include a check after ordinary restarts and source transitions. Confirm that a loop resumes at the intended point, the expected audio track remains selected and no second audio path appears. Monitor stream health after the test as well as alignment. When the file and source are ready, a managed playback arrangement can remove the need to leave your own computer running and watching for a dropped broadcast; StreamNeo turns an uploaded file into a YouTube live stream, but it does not diagnose a sync problem in a source file or guarantee that a correction is right.
Before committing, compare the operating options on the pricing page. When the file and channel are ready, start free — 24-hour trial, no card.
FAQ
Can I fix any audio desync by adding an OBS audio delay?
No. A measured, stable gap may be corrected by adjusting the relevant source timing, but the right direction and amount depend on which part leads. If the gap grows over time, a constant delay only shifts the starting point and does not stop the drift.
Is YouTube stream latency the same as audio and video being out of sync?
No. Latency is the delay between the encoder and the viewer, while sync describes the relative timing of sound and image. Check whether a visible event still matches its sound in playback before changing source timing.
What should I check if sync gets worse during a loop?
Check whether the file is in sync from beginning to end, whether the correct track is selected, and how the source restarts at the loop boundary. Also compare sample-rate settings across the relevant audio path, treating a mismatch as a possible lead rather than a guaranteed cause.
Should I change audio settings while my 24/7 stream is live?
There is no single safe procedure for every encoder and source arrangement. Where possible, reproduce the fault in a controlled test, make one measured change, then verify encoded playback and stream health before changing the production stream.