The Kit and Routine Behind a Reliable Interview Sound Check

When people ask what I carry for a sound check, they usually expect a list of expensive equipment. The honest answer is that the method matters more. A sound check is a repeatable sequence, and the sequence is what catches problems while there is still time to fix them. The same microphones can sound clean at one event and unusable at another, and the difference is almost always whether someone listened carefully before the first question was asked.

In my teaching I describe the sound check as three small questions asked of every speaker: can I hear them clearly, are the levels safe, and is anything else in the room competing with them. If you can answer those three for each person, you have done most of the job. The tools below simply make those answers faster and easier to trust.

Early autumn is a good time to settle on a routine, because many crews are heading back into venues after quieter months and old habits tend to resurface. Writing your sequence down once, in the order you actually do it, saves you from rebuilding it from memory in a hurry on every shoot.

The first tool is a pair of closed-back headphones that you know well. Familiarity matters because you are judging small differences: a faint hum, a slightly boxy voice, a rustle from clothing. If you change headphones from job to job, you spend half the sound check wondering whether the problem is in the room or in your ears. Pick one pair and learn how speech sounds through them.

Where you stand matters too. I try to listen from a spot where I can see the speakers but am not in the middle of the room's noise, such as a corridor door or a ventilation outlet. Monitor at a sensible level, never loud enough to mask a problem. A quiet, steady listening position lets you notice the thing that will otherwise only reveal itself in the edit, when it is far too late to ask anyone to speak again.

I also listen to each microphone on its own before bringing them up together. Solo listening makes a failing cable or a rubbing lavalier obvious, whereas a full mix can hide it behind the other voices.

• Use the same closed-back headphones on every job
• Listen from a spot away from fans and doors
• Solo each microphone before hearing the full mix

The second tool costs nothing: a short, consistent script. I ask each speaker to say their name, then speak for about fifteen seconds at the volume they will actually use. Asking someone to count to ten tends to produce a flat, careful voice that is quieter than their real delivery. A sentence about what they had for breakfast, or how they got to the venue, gets closer to natural speech and gives you a believable level.

While they talk, I watch the meters and listen for the loud moments, not the average. People raise their voices when they get animated, laugh, or lean in to answer a hard question. I set gain with that louder behaviour in mind and leave comfortable headroom, so the first real burst of energy does not distort. If a speaker is very quiet, I would rather move the microphone closer than push the gain high and bring up the room with it.

Keep the script identical for every speaker on a panel. It makes the levels easier to compare and gives the whole recording a consistent starting point.

• Ask for a name and fifteen seconds of natural speech
• Set gain for the loudest likely moment, not the average
• Move the microphone before raising gain on a quiet voice

Beyond headphones, my kit is deliberately plain. The field mixer's own meters show level, and I trust them alongside my ears rather than instead of them. A small notepad is the other tool people underestimate. I write down each speaker's name, which channel they are on and the gain I settled on. If something changes mid-session, or a colleague takes over the desk, that page answers questions that would otherwise cost minutes.

I also keep a rough reference in my head and on paper for what a clean take should look like on the meters in that room. It is not a precise number so much as a pattern: voices peaking comfortably below the top, a steady floor underneath, no flickering from handling noise. When a channel behaves differently from the rest, the pattern tells me to investigate.

None of this needs to be complicated. The aim is a sound check you can repeat on a quiet Tuesday and on a crowded morning with the same result, because the steps do not depend on how rushed the day feels.

The last step of my sound check is the one most often skipped. Once everyone is seated and quiet, I record room tone: thirty to sixty seconds of the space with nobody speaking. Every room has a sound of its own, from air handling to distant traffic to the faint buzz of lighting, and that sound changes whenever a door opens or a heater switches on.

In the edit, room tone fills gaps between cuts so that silences do not suddenly drop to nothing, and it gives the editor a clean sample of the background if they need to reduce it. Without it, they are left borrowing a stray half second from between sentences, which rarely matches. Ask for stillness politely, count it in, and keep your own movements to a minimum while it records.

If the room has several states, such as lights on and a projector running, record tone in the state you will actually film in. A sample taken before the equipment was switched on describes a room that no longer exists.

• Record thirty to sixty seconds with the room quiet
• Capture tone with lights, projectors and fans as they will run
• Stay still and silent yourself during the take

Here is the sequence I follow, in the order I follow it. First I walk the room and note anything noisy. Then I place the microphones and check each one solo through the closed-back headphones. Next I run the short script for every speaker and set gain with headroom. After that I write down channels and levels. Finally, with everyone settled, I record room tone and play back a few seconds of the take to confirm it is clean.

That playback matters. It is the only way to know the recording chain works end to end, from microphone to storage, rather than assuming it does. Five minutes spent listening back has rescued more sessions than any piece of equipment I own.

You can adapt the order to your own setup, but keep it fixed once you have chosen it. A routine only protects you when you do not have to think about it.