How to Fix Audio Sync Issues in Video Podcasts

The lips move. The words follow half a second later. It looks like a badly dubbed foreign film, and it is one of the most common technical problems in video podcasting. This guide explains why it happens, how to fix it in post, and how to stop it from happening again.

Audio sync drift is frustrating precisely because it can persist throughout your recording session and only reveal itself once you are in the edit. You watched the monitor. Everything looked fine. Then you import the footage into DaVinci Resolve or Premiere Pro, and somewhere between the 20- and 30-minute mark, the guest's words stop matching their mouth movements. The longer the recording, the wider the gap gets.

There are four main causes. Each has a different fix. Knowing which one you are dealing with saves hours of guesswork.

A 1960s comic book illustration of a man in a mid-century office, showing panic in duotone pink/magenta style.

Understanding the Three Types of Sync Problems

Before reaching for the offset slider, it helps to know what kind of problem you actually have, because not all sync issues behave the same way.

Fixed offset is when audio and video are out of sync by a consistent amount from start to finish — always a quarter second late, for instance. This is usually a monitoring or hardware latency issue rather than a footage problem, and a single nudge in your editor fixes it cleanly.

Progressive drift is the one that catches most people out. Everything is locked at the start of the recording, then gradually pulls apart over the course of the session. By the end of a 90-minute interview, the audio can be several seconds behind the picture. This is almost always a frame-rate or sample-rate mismatch.

Intermittent desync — where audio snaps in and out of alignment unpredictably — is typically a buffer or USB bandwidth issue, particularly common when routing audio and video through the same device connection.

Cause 1: Variable Frame Rate Footage

This is the single most common cause of drift in podcast editing, and the one most people have never heard of. Most cameras and smartphones record at a variable frame rate (VFR). Rather than capturing exactly 25 frames per second, they capture slightly more or fewer, depending on what is happening in the scene, a technique that reduces file size by allocating more data to fast movement and less to static shots.

The problem is that your editing software locks audio to a fixed timeline. When it tries to marry VFR video to a constant audio track, the two gradually fall out of step. The audio was recorded in real time. The video was not. The result is drift that gets worse the longer the recording runs.

How to check if your footage is variable frame rate

Download MediaInfo (free, available for Mac and Windows). Drag your video file onto it. Look at the Frame Rate Mode field. If it says "Variable", you have VFR footage. If it says "Constant", that is not your problem.

How to check if your footage is variable frame rate

Download MediaInfo (free, Mac and Windows). Drag your video file onto it. Look at the Frame Rate Mode field. If it says "Variable", you have VFR footage. If it says "Constant", that's not your problem.

The fix: convert to constant frame rate before editing

The most reliable solution is to convert VFR footage to constant frame rate (CFR) before it touches your editing software. HandBrake (free) handles this well. Import your file, set the frame rate to "Constant" and choose your target rate — 25fps for UK content, 29.97fps for US distribution. Export, then import the converted file into your editor.

Alternatively, if you use DaVinci Resolve or Premiere Pro, you can place the VFR clip on a timeline and export without any cuts. The export process will render it as CFR, which you can then import into your actual edit project.

Quick Fix — Variable Frame Rate

HandBrake → import clip → set Frame Rate to Constant → choose 25fps (UK) or 29.97fps (US) → export → re-import to editor. Free and takes about as long as the clip itself to process.

Cause 2: Frame Rate Mismatch Between Camera and Timeline

Even with constant frame rate footage, drift happens when your camera and your editing timeline are running at different rates. This is particularly common in multi-camera podcast setups where different cameras are set to different frame rates — one at 25fps, another accidentally left at 29.97fps from a previous shoot, a third at 24fps because someone changed it for a cinematic look.

The numbers look close. 29.97fps and 30fps look like the same thing. They are not. Over an hour of footage, a 29.97fps camera recorded in a 30fps timeline produces 108 extra frames — roughly 3.5 seconds of video that has no matching audio. The audio does not stretch to fill it. It runs out early and everything before the end is slightly wrong.

The 59.94 versus 60fps version of this problem is equally common, and equally invisible until the edit. Blackmagic forum threads are full of people who have spent days troubleshooting what turned out to be a single wrong frame rate setting on the ATEM or in the editing project.

The fix: set everything to the same frame rate before you record

In the UK, 25fps is the standard for online video. Set your cameras to 25fps. Set your ATEM Mini Extreme ISO output to 25fps. Set your editing project to 25fps. Every device in the chain should display the same number. If you are producing primarily for a US or global audience and prefer 30fps, use 30fps throughout — not 29.97 unless you have a specific broadcast reason for it.

If you are already in the edit with mixed frame rates, the cleanest fix is to transcode every clip to your target frame rate before re-importing. DaVinci Resolve's Speed Change panel can reinterpret clips, though this changes the clip duration and should be treated carefully with long-form material.

Cause 3: Sample Rate Mismatch

Audio is measured in sample rate — the number of times per second it captures a snapshot of the sound wave. The two rates you will encounter are 44.1kHz (used for music, CDs) and 48kHz (the broadcast and video production standard). These are not interchangeable, and mixing them causes drift.

Imagine your camera records video at 25fps and captures audio internally at 48kHz. Your external audio interface is set to 44.1kHz. Your editing software project is set to 48kHz. The video and its embedded audio are in sync because they came from the same device. But your external audio — recorded at 44.1kHz and interpreted by a 48kHz timeline — plays back at a slightly wrong speed. Over a one-hour recording, it drifts by roughly 3.6 seconds.

This is a quiet, invisible problem. The audio levels look fine. The waveforms look fine. It is only when you listen carefully, or zoom in at the 45-minute mark, that the drift becomes obvious.

Quick Fix — Sample Rate Mismatch

Set every audio device in your chain — microphone interfaces, mixers, cameras — to 48kHz. Set your editing software's project audio sample rate to 48kHz. This is the video production standard and removes all ambiguity.

Cause 4: USB Buffer Drift (ATEM Mini and Similar Switchers)

If you route both video and audio through a hardware switcher like the ATEM Mini Extreme ISO via USB, you may encounter a specific kind of intermittent or progressive drift that no amount of frame rate matching will solve. The cause is USB buffer inconsistency — the computer is not always pulling video and audio from the USB stream at exactly the same rate, and over a long session the gap accumulates.

This manifests as audio that is locked at the start of a recording, holds for 30 to 60 minutes, then slowly drifts before occasionally snapping back. The Blackmagic forums have extensive threads on this, and the most consistent solutions are to keep the ATEM firmware current, avoid routing audio exclusively through the USB connection for long recordings, and — most reliably — record a separate audio backup through a dedicated audio interface or directly into the camera body.

The ATEM Mini Extreme ISO's strength is its ISO recording capability: each camera gets its own file. For best results, record the primary audio into the camera via XLR or into a dedicated audio recorder running independently, rather than relying solely on the ATEM's USB audio output to a laptop.

The Full Diagnostic Table

Audio Sync Problem Diagnosis Guide
Symptom Most Likely Cause Where to Check Fix Difficulty
Sync is wrong from the very first frame Fixed monitoring latency or hardware delay Playback from the start of the clip Nudge audio offset in editor by a consistent number of frames Easy
Sync is fine at the start, drifts gradually over time Variable frame rate footage or sample rate mismatch MediaInfo — check Frame Rate Mode; check all device sample rates Convert footage to CFR via HandBrake; set all devices to 48kHz Medium
Drift gets worse the longer the recording — several seconds out at the end Frame rate mismatch between camera and editing timeline (29.97 vs 25, etc.) Camera menu and editing project settings Transcode all footage to matching CFR; set project to match Medium
Multi-camera: one angle is fine, another drifts Cameras set to different frame rates Camera settings menus on each device Standardise all cameras to 25fps (UK) before re-recording; transcode existing footage Medium
Sync randomly snaps in and out during playback USB buffer instability (common with ATEM Mini via laptop) Blackmagic ATEM firmware version; USB cable and port quality Update ATEM firmware; use separate audio recorder; switch audio source in OBS/editor to Blackmagic audio device specifically Complex
Separate audio file (from external recorder) doesn't match video at the end Sample rate mismatch between recorder and camera (44.1kHz vs 48kHz) Audio recorder settings; camera audio settings Set recorder to 48kHz; use a clap at the start to manually align Easy
Fine in source file, drifts after export Project frame rate or audio rate set incorrectly in editing software Project settings in DaVinci Resolve or Premiere Pro Match project settings to source footage; re-export Medium
Remote guest footage (Riverside, Zencastr, etc.) drifts against in-studio footage Mixed frame rates from different recording platforms Platform export settings; check each clip in MediaInfo Transcode all remote clips to CFR matching your studio footage before editing Medium

How to Fix It in Your Editing Software

DaVinci Resolve

If the offset is fixed and consistent, right-click the audio clip on the timeline and select Clip Attributes. You can adjust the audio offset here in frames. For drift, go to the Fairlight page and use the Elastic Wave feature if the drift is gradual and non-linear — this lets you manually stretch or compress sections of audio to match the picture without affecting pitch.

For VFR footage causing drift, go to File → Project Settings and confirm that the timeline frame rate matches your primary camera's frame rate. If footage at 29.97 fps is sitting in a 25 fps project, right-click the clip, select Clip Attributes, and change the frame rate interpretation — though transcoding first is always the cleaner approach.

Adobe Premiere Pro

Right-click any audio clip and choose Audio Gain for level control, but for offset, go to the clip's audio properties and adjust Audio Time Unit offset. For larger drift corrections, the Sequence → Sequence Settings menu lets you confirm the project frame rate. If you imported footage before checking, use Modify → Interpret Footage to tell Premiere what frame rate the clip should be treated as — useful when a clip has been wrongly identified on import.

For VFR clips, Premiere is more tolerant than DaVinci Resolve in some cases, but for anything over 30 minutes, converting via HandBrake first is the safer path.

Final Cut Pro

Final Cut handles audio offset through the Inspector panel. Select the clip on your timeline, open the Inspector (press Command+4), and click the Audio tab. You will find a Channel Configuration section with a separate audio offset field, allowing you to nudge the audio forward or backwards in small increments.

For a fixed offset that runs through the whole clip, this is the quickest fix. For progressive drift caused by a frame rate mismatch, Final Cut can reinterpret clip frame rates directly — right-click the clip in the Browser, select Info, and check the frame rate shown against your project settings under File → Project Properties. If they differ, use File → Modify → Retime to conform the clip, or transcode through HandBrake first for anything over 30 minutes. One Final Cut-specific thing to watch: if you recorded on an iPhone or iPad, the footage is almost certainly variable frame rate. Final Cut is more tolerant of VFR than Premiere Pro or DaVinci Resolve, and it often handles it silently, but on recordings longer than an hour, you may still see drift creep in.

The fix is the same as everywhere else: run the file through HandBrake, set it to a constant frame rate of 25 fps before it enters your project.

Prevention: Get the Recording Right

Everything above assumes you are already in the edit with a problem. The better approach is a five-minute pre-session checklist that makes sync issues structurally impossible.

Set a consistent frame rate across every device before each recording session. In the UK, 25fps is the right choice for a video podcast. Check the camera menu. Check the ATEM output setting. Check the editing project template you use. If any of the three says something different from the others, find it now rather than in the edit.

Set every audio device to 48kHz. Camera body audio input. External audio interface. ATEM audio settings. Editing project audio settings. 48kHz is the video production standard — using 44.1kHz (the music standard) anywhere in a video workflow is asking for trouble.

Record a clap or a sharp hand gesture at the start of every take. Even with perfect settings, a visible sync reference at the top of the recording means any residual offset can be snapped into place in seconds. Professional film productions use a clapperboard for exactly this reason. A hand clap costs nothing and takes one second.

If you use an external audio recorder, start it rolling before the cameras and stop it after. This gives you audio that spans the entire video, with no gaps that could cause alignment confusion later.

For multi-camera setups, sync your cameras to a common timecode source if possible. If your cameras do not support timecode, record a reference clap on all cameras simultaneously at the start of the session. This gives your editing software — or you manually — a precise sync point to work from across all angles.

A Note on Remote Guests

Recording remote guests through platforms like Riverside.fm, Zencastr, or SquadCast introduces another variable: the remote footage arrives in its own format and frame rate, often variable, sometimes 30 fps when your studio is running 25 fps. The safest approach is to transcode every remote clip with HandBrake before it enters your editing project — set to a constant frame rate of 25 fps, with audio at 48kHz. This takes a few minutes per file and eliminates the entire category of remote drift problems before they have a chance to appear.

Summary: The Ten-Second Version

  • Fixed offset from the start? Nudge the audio track.

  • Getting worse over time? Frame rate or sample rate mismatch.

  • Check footage with MediaInfo. If it shows a variable frame rate, convert with HandBrake.

  • Set everything to 25fps (UK) and 48kHz consistently before every session.

  • Clap at the start of every recording. Always.

Audio sync problems have a reputation for being mysterious. They are not. They are almost always caused by one of three things — variable frame rate, frame rate mismatch, or sample rate mismatch — and all three are preventable with the same brief checklist run before you press record.

Mark Hunter

Mark is the founder of Postable Limited and the co-founder of the Podcast Studio Glasgow. He became a pioneer of podcasting in 2005 and has worked extensively as a podcast producer, digital marketing consultant and content creator.

https://podcaststudioglasgow.com
Previous
Previous

The Unfair Advantage: Why Every Business Coach and Leadership Expert Needs a Podcast (And Why Your Competitors Won't Have One)

Next
Next

Scottish Podcast Awards: Entry is Now Free (Thanks to Spotify)