Wedding Video Audio: The Frequency Rules, the Consent Problem, and a Redundancy Plan That Holds

Wedding films do not usually fail on picture. They fail on the vows, and the failure is almost always one of three things: a chain that died quietly, a transmitter operating on frequencies that are no longer legal to use, or a microphone that kept recording somewhere it should not have. Record every irreplaceable moment on at least two chains that fail independently, confirm your transmitters tune to bands unlicensed users are actually permitted to use, and disclose recording before you clip a lavalier on somebody and walk away.

Almost every article about wedding audio is a gear roundup. This is the part underneath the gear: the two federal rules that decide which frequencies you may transmit on, the license you are not eligible for, the state statutes that turn a forgotten lavalier into a criminal question, and a redundancy plan built on independence rather than on owning more microphones.

Two rules list every frequency you are allowed to use

Wireless microphones in the United States operate under one of two regimes, and which one you are in is not a choice you get to make.

Unlicensed operation runs under 47 CFR 15.236. That rule permits wireless microphones in the channels allocated to broadcast television, in the 614 to 616 MHz segment of the 600 MHz guard band, and in the 657 to 663 MHz segment of the 600 MHz duplex gap. Power is capped at 50 mW EIRP in the TV band and 20 mW EIRP in the guard band and duplex gap. Operation in the TV band also has to stay outside the protected service contours of co-channel stations, which is the real job your receiver's scan function is doing when you let it pick channels at a venue.

Licensed operation runs under Part 74 as a low power auxiliary station, and 47 CFR 74.802 sets out its frequency list: bands at 26 MHz, 54 to 72 MHz, 76 to 88 MHz, 174 to 216 MHz, 450 to 451 MHz, 455 to 456 MHz, 470 to 488 MHz, 653 to 657 MHz, and the 941 to 952 MHz range.

Now read both lists again and notice what is absent from each. Neither one contains 617 to 652 MHz. Neither one contains 663 to 698 MHz. That spectrum was reallocated to mobile carriers through the FCC's broadcast incentive auction, and wireless microphones had to be out of it by July 13, 2020, as the FCC sets out on its post-incentive auction transition page. A great deal of that equipment was excellent, it still powers on, and it is still being resold secondhand to people entering the industry. Powering on is not the test.

The license that unlocks the other list is closed to you

The obvious response to the paragraph above is to get licensed. You cannot, and the reason is worth knowing precisely, because it quietly sets the ceiling on what a wedding audio rig is allowed to be.

47 CFR 74.832 lists who may hold a low power auxiliary station license: broadcast and low power TV licensees, broadcast network entities, cable operators producing program material, motion picture producers, television program producers, Broadband Radio Service licensees, large venue owners and operators, and professional sound companies. The last two categories are the only plausible doors for an independent filmmaker, and both come with a certification attached. The applicant has to certify that it routinely uses 50 or more low power auxiliary station devices, where the use of those devices is an integral part of major events or productions.

A wedding videographer with four transmitters and a spare is not that applicant, and nothing about growing a studio changes it. So the practical position is fixed: you are a Part 15 operator permanently. That means the 653 to 657 MHz duplex gap segment reserved for licensed users is not yours, and the power limits above are your limits, not a conservative default you can dial past.

Two consequences follow that are worth designing around rather than resenting. The first is that range is a rigging problem, not a wattage problem. At 50 mW, line of sight and antenna placement do more for you than any menu setting, which is an argument for a receiver near the altar rather than on a camera forty metres back. The second is that systems working in the 2.4 GHz band, which is where a lot of modern compact kit lives, are ordinary unlicensed devices sharing spectrum with every phone, hotspot and venue access point in the room. They are perfectly legitimate and often the right call, and they are also sharing a band with two hundred guests who all just connected to the wifi. Neither approach is safe on its own, which is the whole argument of the next section but one.

The ceremony itself is not the exposure. A ceremony conducted in front of a hundred and twenty seated guests is not a private conversation by any reading, and the couple hired you to record it. The exposure is the four hours on either side, and it comes from the single most standard practice in wedding audio: clip a transmitter on early, set it, and forget it.

Several states treat recording a private conversation without the agreement of everyone in it as a crime, and they include some of the largest wedding markets in the country:

That Massachusetts framing is the useful one for our craft, because it names the thing that actually distinguishes a professional audio rig from a bug. The offense there is secrecy. A transmitter someone watched you clip on, was told is recording, and can mute is not secret. A transmitter still running two hours later, in a room you left, on a person who has stopped thinking about it, has drifted toward the other definition without anybody deciding that it should.

You can feel exactly where this goes wrong. You are in the edit on a Tuesday, working through the groom's morning, and the timeline keeps running past the point where you picked up your camera and walked out. Ten minutes of a room tone, then his father sits down next to him and says something that is plainly not for you, and neither of them has any idea the little box is still live. Nothing bad will come of it. You will delete it. But you are now the only person who heard it, you did not mean to be, and there was a version of the morning where you simply said out loud that the mic stays hot until you take it off him.

State rules vary considerably and change, and the four above are examples rather than a complete list. The Reporters Committee maintains a state-by-state recording guide that is the fastest way to check a state you are travelling to before you take the booking.

Redundancy is about independence, not microphone count

Here is the framework we use, and the first rule is that a chain is only redundant if it can fail without taking the other one with it. Two lavaliers feeding one receiver are one chain. Two transmitters on the same battery type you changed at the same time are closer to one chain than two. A camera-mounted shotgun and an in-camera recording on the same body are one chain, because the body is the shared failure.

Four capture points, each owning something different:

Now the arithmetic, and read the assumption carefully because we are not claiming to have measured this. Suppose any single chain has a one in twenty chance of producing something unusable at a given wedding: a cell that was not as fresh as you thought, a lavalier buried under a boutonniere, gain set for a speaker and then handed to a whisperer, an RF hit at the worst nine seconds. Five percent is a made-up rate chosen to be plausible, not a statistic.

One chain fails on one wedding in twenty. Two chains that fail independently both fail on one wedding in four hundred. Three, one in eight thousand. The precise numbers are worthless and the exponent is the entire point: each genuinely independent chain multiplies your exposure down, and each dependent one you add does almost nothing. This is why the shooter with three transmitters on one receiver and the shooter with two transmitters and a board feed are not in the same business, even though the first is carrying more equipment.

Three habits do more than any purchase:

What to write down, and where

Audio sits in a different contractual category from the other technical risks in a wedding film, and it is worth being clear about why. Aerial coverage is a genuinely conditional deliverable, and we argue in the drone rules post that it should be sold and written as one, refundable if weather or airspace or venue policy prevents the flight. Audio is not conditional. Nobody books a film on the understanding that the vows are optional. So the clause you need is not an escape hatch, it is a description of the standard of care you actually meet plus a proportionate limit on liability.

The pattern here is the same one that runs through music licensing and the drone rules, and it is becoming the through line of this whole series. The most emotionally load-bearing parts of a wedding film are governed by rules that the person selling the film has usually never read. In every case the compliance is cheap in advance, the failure is expensive after delivery, and the version of you who checked in February is protecting the version standing at the back of a church in September.

None of this argues for spending more. The rig described above is not an expensive rig, and the frequency check costs an afternoon once. What it argues for is that the audio side of a videography package is where the actual professional risk sits, long after the camera question has stopped being interesting.


Common questions

What is the best way to record audio at a wedding?
Capture every irreplaceable moment on at least two chains that fail independently. In practice that means a transmitter on the officiant and a second on one partner, so the vows exist twice from different bodies, plus a recorder taking a feed from the house or DJ desk for the speeches and a camera-mounted shotgun for room tone and sync. Two microphones running into the same receiver is one chain, not two. The property that matters is independence, not microphone count.
Are 600 MHz wireless microphones still legal to use?
No. Look at the two rules that list permitted frequencies. Unlicensed wireless microphones under 47 CFR 15.236 may operate in the TV broadcast channels, the 614 to 616 MHz guard band and the 657 to 663 MHz duplex gap. Licensed low power auxiliary stations under 47 CFR 74.802 get their own list, which tops out around 653 to 657 MHz in that region. Neither rule contains 617 to 652 MHz or 663 to 698 MHz, because that spectrum was auctioned to mobile carriers and wireless microphones had to be out of it by July 13, 2020. Plenty of that gear is still circulating on the used market.
Do wedding videographers need an FCC license for wireless microphones?
You do not need one, and more to the point you cannot have one. A low power auxiliary station license under 47 CFR 74.832 is limited to broadcasters, cable and program producers, large venue operators and professional sound companies, and the last two categories must certify that they routinely use 50 or more such devices as an integral part of major events or productions. A wedding videographer running four transmitters does not qualify. You operate as an unlicensed Part 15 user, which caps you at 50 mW EIRP in the TV band and 20 mW in the guard band and duplex gap.
Is it legal to leave a lavalier microphone recording at a wedding?
During the ceremony, generally yes, because a ceremony in front of assembled guests is not a private conversation. The risk sits in the hours around it. A transmitter clipped on at nine in the morning that keeps recording while the wearer is alone with a parent is capturing a private conversation, and several states treat that as a criminal matter. California requires the consent of all parties to a confidential communication, Florida makes unlawful interception a third degree felony, Illinois prohibits surreptitious recording of a private conversation, and Massachusetts turns on secrecy rather than consent, with penalties up to $10,000 and five years. Disclose the recording, make the mute obvious, and power down between blocks.
How many audio sources should I run at a wedding?
Three independent chains on the ceremony and two on the speeches is a sensible working floor. The reason is arithmetic rather than paranoia. If any single chain has a one in twenty chance of producing something unusable, one chain fails on five percent of weddings, two independent chains fail together on one in four hundred, and three on one in eight thousand. The exact failure rate is unknowable, but the exponent is real, and audio is the only part of a wedding film that cannot be recovered in post.