Speeches and toasts are the part of a wedding day that guests quote for years, and the part most likely to be ruined by a single bad microphone decision. A father's voice cracks on the third line, the best man's joke lands, the room laughs, and a bridesmaid's phone buzzes in the wrong pocket. If you are a wedding videographer who outsources editing, the speeches are also where the gap between "we filmed it" and "we can actually cut it" shows up most clearly.
This guide covers the full chain for reception speeches: what to offer clients, what to ask before the day, how to capture clean audio and usable pictures, what to hand your editor, how a good editor shapes a speech, and how to sell and quality-check the result. It assumes you already film multicam and already send footage to an outside editor. If you do not yet, start with the basics of handing wedding footage to an editor.
Speeches are the only part of the day where words, not images, carry the story. In a highlight film, a cutaway to a rolling-eyed groomsman can be swapped for any other cutaway. In a speech, the sentence itself is the content, and if the audio is unusable the moment is gone. No colour grade fixes that.
They are also long and unpredictable. A ceremony has a script and a rehearsal. A reception speech might run three minutes or twelve, include a spontaneous heckle, a dropped glass, or a guest who grabs the microphone uninvited. Your capture plan has to survive all of that without a retake.
There are three practical reasons to treat speeches as a separate workflow rather than "just more reception footage":
- Audio dominates. A mediocre picture with clear speech is watchable. Perfect picture with muffled speech is not.
- Editing decisions are editorial, not just technical. What you cut from a speech changes what a person appears to have said. That needs a brief, not a guess.
- Clients value them differently. Some couples want every word of every toast; others want a ninety-second distillation inside the highlight film. You need to know which before the day.
The rest of this article walks through each stage in the order it happens: offer, plan, capture, hand off, edit, price, check.
Decide what you are delivering: three speech products
Before you plan capture, decide what the speeches become. There are three common products, and they ask very different things of your editor.
| Product | What the client gets | Editing effort | Capture demands |
|---|---|---|---|
| Speech moments inside the highlight film | Two or three short lines from the best speeches woven into the main film | Low: select and place | Clean audio on the lines that matter |
| Speeches edit | A separate film of the speeches, trimmed for pacing, with reaction cutaways | Medium to high: trimming, cutaways, audio cleanup | Two angles plus clean primary audio |
| Full-length speeches | Every speech, uncut or nearly uncut, one file per speaker | Low to medium: sync, audio cleanup, titles | Continuous coverage and reliable backup audio |
The most common mistake is promising "the speeches" without saying which of these three. A couple who imagines every word and a videographer who planned two sound bites will both be disappointed. Write the product into the package description, and say plainly whether the full-length version is included, an add-on, or not offered.
If you also deliver a cut for social media, remember that speech clips there are usually short and captioned, which changes what you ask the editor to mark during the edit. That is a separate decision from the three products above.
Plan the speeches before the day
Most speech problems are information problems. The videographer did not know who would speak, in what order, from where, or into what. Settle those facts in advance, ideally through your client questionnaire and a conversation with the venue or planner.
The questions to ask
| Question | Who to ask | Why it matters |
|---|---|---|
| Who is speaking, and in what order? | Couple or planner | Lets you label files and prepare angles for each speaker |
| Will speakers use a microphone, and whose? | Venue, DJ or band | Decides whether you can take a feed from their mixer |
| Where will speakers stand? | Venue or planner | Decides camera positions and the cable or wireless route |
| Is there a hard time limit or dinner service in the middle? | Planner or caterer | Affects battery, card space and when you can swap media |
| Are speeches before, during or after the meal? | Planner | Affects light levels and where guests are seated |
| Is anyone likely to speak who is not on the list? | Couple | Tells you whether to keep recording between formal speeches |
| Does the couple want full-length speeches, or a short edit? | Couple | Decides the product and the effort |
On timing, published wedding-planning guidance gives a rough sense of scale. Bridebook's guidance suggests that individual speeches commonly run around five to seven minutes, that the whole speech portion usually lasts twenty to thirty minutes, and that it should not exceed forty. Treat those as planning figures, not promises; real speeches overrun. Build your media and battery plan around the longer end, not the average.
Speeches traditionally come after the main course and before dessert, though they can follow dessert as well. That matters for light: a reception room that was bright during the meal may dim as the band or DJ prepares, and your camera settings should anticipate it.
Finally, ask whether anyone is uncomfortable being filmed or recorded. Speeches can include private family material. If the couple wants a say over what is shared online, agree that before the day and note it in your brief to the editor.
Capture clean speech audio
Audio is the single most important input. Plan at least two independent sources for every speech, so that a dead battery, a dropout or a dull microphone does not erase the only copy of a father's toast.
| Audio source | Strength | Weakness | Role |
|---|---|---|---|
| Feed from the venue's or DJ's mixer | Usually the cleanest voice, close to the speaker's microphone | Depends on the venue's gear and operator; may include music or announcements | Primary when available |
| Wireless lavalier on the speaker | Consistent distance to the mouth; independent of the venue | Needs a few minutes with the speaker and a hidden placement; clothing rustle | Primary or strong second |
| Shotgun microphone on the camera | Always available; also records reactions and laughter | Captures room noise and varying speaker distance | Backup and ambience |
| Portable recorder near the speaker or at the head table | Independent recording that cannot be broken by a wireless dropout | Must be started and monitored; placement is limited | Backup |
Take a feed when you can
If the venue or DJ has sound reinforcement, ask for a feed from their mixer and record it separately. Photography and video trade guidance on wedding audio recommends exactly this: if the room is amplified, record a line out from the soundboard, and if it is not, aim a shotgun microphone from the camera at whoever is speaking. A mixer feed often carries the cleanest version of a speaker's voice because it comes from the microphone they are actually holding.
Ask early. Day-of requests are harder to meet, and not every operator is happy to hand over a feed. When you do take one, record it on a device you control and label the file immediately so no one confuses it with ambient audio later.
Use a lavalier where the speaker is willing
A lavalier on a toast-giver is the most consistent option, because the microphone stays the same distance from the mouth wherever the person walks. It does need a short conversation: you are asking a nervous speaker to wear a transmitter. Hide it under clothing if possible, because visible clip-ons can distract in the picture and may not be welcome with the bridal party. Keep the ask short and polite, do it before the speeches begin rather than mid-event, and always have a second source running in case the speaker declines or forgets to switch it on.
Many compact wireless systems also record internally to the transmitter. The Rode Wireless GO II, for example, is documented as recording on board so that you keep a backup if the signal drops, and it includes a safety channel. Check the documentation for whatever kit you own, and test the recording backup before wedding season rather than discovering during a toast that it was not enabled.
Set levels with headroom
Speeches have sudden peaks: laughter, applause, a glass tapped near the microphone. One widely read wedding-industry audio guide describes keeping levels around minus six to minus twelve decibels, so that loud moments do not clip. A quiet recording can be lifted in editing; a distorted one cannot be repaired. Err on the side of lower and clean.
Record a few seconds of room tone
Before the first speech, record around ten to thirty seconds of the room with no one talking. Editors use room tone to fill gaps after trimming, so a cut does not leave an unnatural silence. It takes almost no effort and saves real trouble later.
Test and monitor
Do a short check with the first speaker, and glance at the meters or listen on headphones during each speech if you can. If you are shooting alone, a small second camera or recorder left running is better than a clever setup you cannot watch. For a deeper look at preparing and handing over sound, see how to prep wedding audio before you outsource editing.
Plan for the unplanned speaker
Almost every reception has at least one unscheduled speaker: an uncle who takes the microphone after the formal toasts, a friend who wants to say a few words, or a child who is pushed forward by the table. These moments are often the most memorable, and they rarely arrive near your lavalier or your mixer feed. Keep one camera ready on a wide frame, keep the shotgun microphone live, and keep your recorder running throughout the speech block. If you have to choose between a tidy setup and being ready for a surprise, choose being ready.
Also decide in advance who is responsible for what if you work with a second shooter. One person owns audio and monitors the sources; the other owns reactions. When both of you chase the same shot, the edit ends up with two copies of one angle and nothing on the people who were crying.
Cover speeches with the right pictures
Good speech coverage gives the editor choices. The goal is not a beautiful single shot but enough variety that a trimmed speech can still look natural.
| Angle | What it shows | Why the editor needs it |
|---|---|---|
| Wide, locked off | Speaker, head table and room | A safe base layer that runs for the whole speech |
| Tight on the speaker | Face and expression | The emotional core; usually the main picture |
| Reaction shots | Partner, parents, grandparents, friends laughing or tearing up | Cutaways that hide trims and add emotion |
| Detail and atmosphere | Glasses raised, hands, candles, table | Texture for transitions and speech moments in the highlight film |
If you have a second shooter, give them the reaction job and the second angle on the speaker. If you are solo, run one camera wide and locked, and use the other to move between the speaker and the reactions. A locked wide shot is the backbone that lets the editor build a continuous speech even when you are reframing on the other camera.
Keep running during pauses
Do not stop recording between speakers. The best reactions often come while someone is being introduced, and spontaneous guest toasts tend to arrive without warning. Stopping and starting for every speaker also risks missing the first sentence of the next one. Manage your media so that long takes are safe, not so that you can stop frequently.
Match settings across angles
Use consistent frame rates and shutter behaviour on every camera covering a speech so that the editor can switch between angles without visible changes in motion. If you sync with a clap or an audio reference, make sure every camera records the same reference audio. The guide to syncing multicam wedding footage explains how to make that painless for whoever builds the multicam clip.
Hand off speeches so the editor does not guess
A speech is one of the hardest things for an editor to organise from a loose folder of clips. They need to know who is speaking, which file is which, which audio is the best, and what you promised the client. Put that in a short written brief, and organise the footage so it matches.
| Item to include | Example | Why |
|---|---|---|
| Running order | 1 Father of the bride, 2 Groom, 3 Best man, 4 Surprise toast from sister | Lets the editor label the timeline correctly |
| Speaker names and relationships | "Mark, best man, groom's brother" | For titles and for understanding references |
| Best audio source per speech | "Mixer feed best for speeches 1 and 2; lav only for speech 3" | Saves the editor from auditioning every track |
| Product promised | "Full-length speeches plus a 4-minute speeches edit" | Sets scope and avoids rework |
| Sensitive material | "Do not use the story at 3:20 in speech 2" | Protects the couple and the speaker |
| Known problems | "Lav dropped out for 20 seconds during speech 3" | Directs the editor to the backup source |
Name and organise the files
Put each speech's footage and audio in a clearly named folder, in running order, with the audio sources kept separate and labelled. A name like Speech-03-BestMan-MixerFeed is more useful than a camera-generated string of numbers. Keep the room tone file in the same folder. If you add timestamps for the moments the couple will probably want in the highlight film, the editor can find them quickly. The same discipline you use for any footage hand-off applies here, just with more attention to audio.
Say what "good enough" means
If one speaker was inaudible on a bad day, tell the editor what you want: salvage as much as possible, subtitle the unclear sections, or leave that speech out of the edited film. It is easier to agree on that in advance than to explain it in revision rounds. If you want to control how many rounds your client can request, see how to structure revision rounds before outsourcing.
How a good editor shapes a speech
Editing a speech is closer to documentary editing than to highlight cutting. The editor's job is to preserve the speaker's voice and intent while respecting the audience's attention. Understanding what they do helps you brief them well.
Syncing and audio first
An editor will usually sync all cameras and audio sources to the best audio track, choose that track for the master, and clean it. Cleaning can mean reducing steady room noise, controlling clipping, balancing levels, and, where the tools allow, using voice-isolation or speech-enhancement features in the editing software. These features are useful on moderate background noise and unhelpful on badly distorted recordings, so treat them as repair, not a substitute for good capture.
Trimming rules
When a speech runs long, a good editor trims without changing meaning. Typical choices include:
- Removing long pauses, false starts and repeated phrases that do not add emotion.
- Cutting digressions that only the immediate family would follow, if the product is a shortened edit.
- Keeping the first sentence, the clearest story, and the final line, because those give the speech its shape.
- Leaving laughter and reaction in place, because they are part of the content.
Every cut inside a continuous shot creates a visible jump. Editors usually hide it with a cutaway to a reaction, a change of angle or a detail shot, which is why your coverage plan matters. Without alternate angles, a trimmed speech can look choppy.
Length targets for edited speeches
| Deliverable | Sensible approach |
|---|---|
| Speech moments in the highlight film | One or two short lines per speaker, chosen for clarity and emotion |
| Speeches edit | Trimmed so that each speech keeps its story, with the total length agreed with the client |
| Full-length speeches | Minimal trimming; remove dead air at the start and end only |
There is no universal correct length, and the right answer depends on how long the speeches ran and what the client wants. Agree the speeches edit's length in writing.
Titles and captions
A simple lower-third with the speaker's name and relationship helps viewers who were not at the wedding, and lets the film live on after the day. Subtitles are worth considering for any section with unclear audio or for guests who may watch with the sound off. Ask your editor what they can supply and agree the style in advance.
Work with the editor, not around them
The best speech edits come from a short conversation before the work starts. Tell the editor what the couple said about the speeches in the questionnaire, which relatives matter most, and whether humour or sentiment should lead. A joke that works for the room can read oddly to an editor who does not know the family, and a quiet line from a grandparent can matter far more than its volume suggests. When the editor understands who the speakers are, trimming decisions improve and revision rounds shrink.
If the first cut comes back and a speech feels wrong, be specific. Instead of saying a speech feels too long, name the timecode of the section you want shortened and say why. Editors can act on a precise note in minutes; a vague one starts another round.
Price and sell the speeches
Speech editing takes real time: syncing, audio repair, trimming, cutaways, titles and review. Treat it as its own line item rather than something absorbed into "reception coverage". The exact figures depend on your market, your editor's rates and the length of the footage, so use your own numbers rather than copying someone else's.
| Approach | When it fits | Risk |
|---|---|---|
| Include speech moments only | Entry-level packages | Clients may later ask for full speeches |
| Offer full speeches as an add-on | Mid-range packages | Needs clear rules on audio quality |
| Include a speeches edit in premium packages | Higher-end packages | Must account for editing time in your price |
| Charge per speech or per minute of footage | When length varies widely | Needs a clear estimate before the day |
Whichever approach you choose, state three things in your contract or package description: whether speeches are guaranteed, what happens if the venue's audio is unusable, and what the client may and may not ask for later. Speeches depend on factors you do not control, such as a quiet speaker or a venue that refuses a mixer feed, so avoid guaranteeing a result you cannot ensure. Promise your process and your backup plan.
When you outsource, ask your editor how they quote speech work. Some charge by the finished minute, some by the hour and some include a standard speech package. Compare that with what you charge the client, because the add-on needs to cover the editor's cost plus your own time, not just look attractive on a price list. A brief on what the client gets, set out before the wedding, also makes your outsourced workflow easier to manage.
Quality-check the speeches before delivery
Run a dedicated check on speeches, separate from your general review of the film. Errors here are personal: a misspelled name or a clipped sentence in a father's toast is noticed instantly.
| Check | What to look for |
|---|---|
| Names and titles | Correct spelling, correct relationship, correct order |
| Audio clarity | No clipping, no sudden level jumps, no dropouts |
| Sync | Lips match voice on every angle |
| Cuts | No change in meaning, no half-words at edit points |
| Cutaways | Reactions match what is being said at that moment |
| Sensitive content | Nothing removed from the brief appears in the final edit |
| Beginning and end | Speeches start and finish cleanly, with no abrupt cut mid-sentence |
Listen on both headphones and a laptop speaker; problems that vanish on one often show on the other. If you use a review platform, put time-coded notes against specific moments rather than describing them in an email. When your editor returns the first cut, check the speeches against the original audio, not only against the edit, to confirm nothing important was lost.
Common mistakes to avoid
- Relying on one audio source. Always have a second, even if it is only the camera microphone.
- Assuming the DJ's feed will be available. Ask in advance and prepare a fallback.
- Stopping recording between speakers. Reactions and impromptu toasts happen in those gaps.
- Not briefing the editor. Labels and a short note save hours.
- Over-trimming. A shortened speech that no longer sounds like the speaker disappoints everyone.
- Promising what you cannot control. Guarantee your process, not the venue's microphone.
Frequently asked questions
How many audio sources do I need for wedding speeches?
At least two independent sources for every speech. A common combination is a feed from the venue's mixer or a lavalier on the speaker as the primary source, plus a camera-mounted shotgun microphone or a separate recorder as backup. Two sources protect you against battery failures, wireless dropouts and operator mistakes, which are the usual causes of lost speech audio.
Should I film the full speeches or only highlights?
Record the full speeches whenever you can, even if the client only buys a short edit. Recording costs little once the cameras are in place, while missing a speech cannot be fixed later. What you deliver is a separate decision: speech moments in the highlight film, a trimmed speeches edit, or full-length files. Agree this with the couple in writing before the day.
Can AI speech enhancement rescue a bad recording?
It can help with moderate background noise and some room echo, but it cannot restore words lost to heavy clipping or a missing signal. Tools in editing software that clean up speech work best on recordings that were already reasonably clear. Treat enhancement as a repair step for small problems and rely on good capture and backup audio for the rest.
How long should an edited wedding speech be?
There is no fixed rule, and it depends on how long the speaker spoke and what the client wants. Published planning guidance suggests that individual speeches often run around five to seven minutes, so a trimmed edit might keep the strongest portion of that while preserving the speaker's story. Agree the total length of the speeches edit with the client and your editor before work begins.
What if the venue will not let me take a feed from the mixer?
Fall back to a lavalier on the speaker and a camera-mounted shotgun or portable recorder placed close to the speaking position. Tell your editor which source is primary and which is backup, and set the client's expectations: speech audio quality depends partly on the room. Having a documented backup plan is more useful than a promise you cannot keep.
Do I need permission to share speeches online?
Speakers may say private things, so it is sensible to agree with the couple what can be shared publicly. Many videographers deliver full speeches privately and use only short, approved lines in public social posts. Put your policy in your contract and ask the couple to confirm before posting anything beyond the private delivery.