How an Audio Sync Audit Works: What We Trace and What We Write Down

Why we are describing the audit now

Over the past few weeks several studios have asked us what an Audio Sync Audit involves before they commit to one, so I want to set it out in one place. The service is simple to state: we trace lip-sync and latency problems across audio and video chains and document the fixes. The detail of how that happens is less obvious from the outside, and engineers rightly want to know how much of their time we will need and what they get back.

This note covers the audit as we carry it out here in Cambridge, with Priya Nair leading the measurement work and me contributing where camera and display paths are involved. It does not cover the seasonal checks worth doing before winter schedules begin, which another article handles, nor the wider workflow review, which is a separate service.

Where we begin: the symptom, not the equipment

We start with a conversation about what people actually see and hear. A presenter whose mouth seems slightly ahead of the voice on the programme output, a guest who drifts out of sync only on the remote feed, a recording that is fine in the control room but wrong on playback: each points to a different part of the chain, and the description narrows our search considerably before anyone touches a cable.

We ask for a signal flow diagram if one exists, and if it does not we draw one with your engineers on site. That drawing is often useful in itself, because many studios find that the chain as built differs from the chain as remembered. We note every device that holds, buffers or reclocks signal between microphone and output.

• The symptom described in the operators' own words
• Which outputs show the fault and which do not
• Any recent change to equipment, firmware or routing

Measuring the chain stage by stage

Once the diagram is agreed, we measure delay at each stage rather than only at the end. A single end-to-end figure tells you that something is wrong but not where. By injecting a reference signal and observing it at successive points, we can see which device adds time and whether that time is fixed or varies with format, frame rate or load.

Video processing is the usual source of added delay: scalers, frame synchronisers, converters and some monitors all buffer frames. Audio paths can add their own, particularly digital mixers with sample rate conversion or embedded audio that travels through video equipment. We record each figure and the conditions under which it was taken, so that someone repeating the test later can compare like with like.

• Delay measured per device, not just end to end
• Readings repeated at the formats you actually broadcast in
• Fixed offsets separated from drifting ones

What tends to turn up

Most faults we find are ordinary. A processing stage added during a refit without compensating audio delay is common. So is a monitor or recorder fed from a different point in the chain than the programme output, which makes the fault look intermittent when it is only inconsistent. Occasionally the cause is a mismatch of frame rates between a remote contribution and the studio, which shows up as slow drift rather than a constant offset.

We do not assume the equipment is at fault. Sometimes the correct answer is that the chain is within a tolerance most viewers will not notice, and the complaint comes from a monitoring position that is wrongly set up. Saying so plainly saves a studio from replacing hardware it does not need to replace.

The written findings you receive

The output of the audit is a document your engineers can act on without us in the room. It sets out the signal flow as measured, the delay at each stage, the cause or causes we identified, and the fixes in the order we would apply them. Where a fix means a setting change, we give the setting and the value. Where it means a hardware or routing change, we describe it and say what we would check afterwards.

We also include a short section on how to verify the result, so that after the work you can confirm for yourselves that the offset has gone. Priya reviews every report before it leaves us, and Helen Marsden reads it for clarity, since a finding nobody can follow does not help a night-shift engineer.

• Measured signal flow with each delay recorded
• Causes named and fixes listed in order
• A method for checking the result afterwards

What we ask of you beforehand

To keep the visit short we ask for a few things in advance: any existing diagrams, the formats and frame rates you work in, a note of when the fault was first noticed, and access to the equipment at a time when it can be taken off air. A short window with the chain free matters more than a long one with interruptions, because measurements taken while a show is running are hard to trust.

If you are unsure whether your problem is a sync fault at all, send us a description through the contact page. We would rather tell you honestly that a different service suits you better, or that a small check by your own team will settle it, than book an audit that is not needed.

An Audio Sync Audit is a methodical trace from microphone to output, ending in a document your own engineers can use. We measure each stage, name the cause plainly and set the fixes in a sensible order. If lip-sync or latency has been troubling your studio, tell us what you see and hear through the contact page, and we will explain what an audit would involve for your chain.