Guide
How to edit a talking head video in four steps, in the right order
Updated
You edit a talking head video in four steps: cut the pauses and ums, add captions, show what you talk about, then add music and an ending. Here is how to do each step by hand, and what an AI editor does for you.
In short
- You edit a talking head video in four steps: cut the ums and long pauses, add captions, show what you talk about, then add music and a clear ending.
- CapCut can write your captions and Descript can remove ums, but you still make the other cuts and place every screenshot yourself.
- In Cutroom's Magic Editor, you press one button and Jacob, its AI editor, does the four steps in its first edit.
- You change Jacob's edit by asking in plain words, such as “Cut the first sentence” or “Turn the music down a bit”.
- Jacob's first edit costs 250 credits, a chat change costs 40, and a question that changes nothing is free.
How do you edit a talking head video?
You edit it in four steps: cut, add captions, show what you talk about, then add music and an ending. Do them in that order. If you cut after the captions and screenshots are in, you have to move them again.
The drawing below shows the four steps done in one edit by Jacob, the AI editor in Cutroom's Magic Editor. Before it starts, you pick one of four styles: Steps, Skit, Verdict or Poster. The style sets how captions and pictures look.
- Cut the ums, false starts, repeats and long pauses, so every sentence starts clean.
- Add captions that are big, short and correct, so people can follow with the sound off.
- Put a screenshot, a screen recording or a stock clip on the words where you talk about it.
- Add quiet music, check the sound level and end on one clear next step.
A bookkeeper's app comparison, edited in four steps
Jacob's first edit does the four steps. Then she asks for three changes, and each one makes a new version.
Drawing of an example clip, not a customer video
Before · the clip as filmed
A 70-second phone video of a bookkeeper comparing two invoicing apps
“So, um, I get asked this a lot. … (pause) I tried both invoicing apps for a month, and the, the second one is uh the one I kept.”
struck through = what Jacob cuts
After · Jacob's first edit, Verdict style
“I get asked this a lot. I tried both invoicing apps for a month, and the second one is the one I kept.”
- On “apps”: her screenshots of both apps on a laptop, with her in a small box at the bottom
- On “kept”: a typed call to action, then a closing screen asking people to follow her
- Under the whole video: captions that follow her voice, set in the style's own font
- Under the whole video: calm music under her voice, with the sound levelled
Then you ask Jacob
- “Cut the first sentence”v2the same edit, now starting on “I tried both invoicing apps”
- “Make the key words stand out more”v3the same edit with more of her key words standing out
- “Turn the music down a bit”v4the same edit with the music quieter under her voice
Before you edit: film it so the edit is easy
Most editing problems start while you film. A few habits save you work in every step below.
- Film upright, because a vertical video crops the sides of a wide clip.
- Film somewhere quiet, because noise is hard to clean up later, and the Magic Editor doesn't remove it.
- If you trip over a line, pause and say the whole sentence again instead of stopping the recording.
- Say your strongest line early, because a sentence can be cut but not moved to the start.
- Look at the lens, not at yourself on the screen.
How Cutroom's Magic Editor edits a video like this
Jacob is the AI editor inside Cutroom's Magic Editor: it edits your video for you, then changes it when you ask in a chat.
The 4 styles, as example edits
Steps
Three steps, four angles
Skit
Hook, hard stop, demo card
Verdict
The tester's verdict
Poster
Typography does the talking
Hover or tap an example to play it, without sound. These are example edits of each style; the presenters in them are generated with AI.
- Your take: one video of you talking, up to 3 minutes
- Your style: one of 4: Steps, Skit, Verdict or Poster
- Your files: screenshots, a screen recording or a logo, if you have them
- Your goal: what the video is for, in a few words, if you like
- Jacob makes it: normally 10 to 15 minutes, and you get an email
- Your edit: ask Jacob for changes in a chat, then export
Then you ask Jacob, for example
- “Make this about 10 seconds shorter”
- “Put a clip on the strongest sentence”
- “Change the music to something calmer”
Each change is a new version, and going back is free. First cut 250 credits, each change 40, questions free, export 30 credits per started minute.
Step 1: cut the pauses, ums and false starts
Play the clip and cut each um, false start and repeated line. Then shorten the long pauses, but leave a short breath between sentences so you don't sound rushed.
Jacob does this in the first edit, and each of the four styles shortens long pauses. If you want it tighter or slower, you ask for a shorter video or to keep your pauses.
- Descript can find the ums and uhs in an English transcript and remove them for you, and you check the rest by ear.
- If you said a line twice, keep the better one and cut the other.
- In the Magic Editor, you ask for cuts in words, such as “Remove all the ums and uhs” or “Make this about 10 seconds shorter”.
- If your first sentence is a warm-up, ask “Cut the first sentence”, and the video starts on your point.
- If the edit feels rushed, ask “Keep my pauses, it feels too rushed”.
- You can also tap words in the transcript to cut them or bring them back, which is free.
Step 2: add captions people can read
Generate captions from your voice, then read every line and fix names, brand words and numbers. Keep them short and big enough to read on a phone.
Jacob does this in the first edit, in the language you speak. It uses the caption look of the style you picked: Steps, Skit, Verdict or Poster.
- In the CapCut app, you tap Captions, then Auto Captions, confirm the language and tap Generate, then fix any line.
- In the Magic Editor, you fix a misheard word by tapping it. The captions change, and your voice stays as it is.
- You ask for caption changes in words, such as “Make the key words stand out more” or “Use less text on screen”.
- The style you picked sets the fonts, so you can't choose your own in the Magic Editor.
- There's no captions file to download, because the captions are part of the video.
Step 3: show what you talk about
When you mention a screen, a product or a result, show it on those words. Use your own screenshots and recordings first, and stock clips only for the gaps.
Jacob does this in the first edit when you upload files, and finds b-roll for the rest.
- By hand in the CapCut app, you tap Overlay, pick the screenshot and tap Add overlay, then trim it to the words where you mention it.
- In the Magic Editor, a screenshot shows while you talk about it, and a screen recording plays as the demo.
- In the Magic Editor, a product photo appears on the word where you name the product, and your logo closes the video.
- Jacob looks for b-roll in your own clip library first, then on the stock sites Pexels and Pixabay, and puts each clip on the words it belongs to.
- You ask for more in words, such as “Can you find 3 b-roll clips that fit what I'm saying?” or “Put a clip on the strongest sentence”.
- The clips are never AI-generated. Each one is your own file, a clip from your library or stock.
Step 4: music, sound and the ending
Put quiet music under your voice and check that you're easy to hear over it. End on one clear next step, such as a link to tap or a word to comment.
Jacob adds music and sound effects in the first edit. The video ends on your call to action or your logo, if you give one.
- By hand, put a music track under your clip and lower it until every word is easy to hear, then check on phone speakers.
- Whichever of the four styles you pick, Jacob adds sound effects, and you can ask it to make them quieter or turn them off.
- You ask for music changes in words, such as “Change the music to something calmer” or “Turn the music off”.
- You can change the call to action in the chat, such as “Change the call to action to Try it free”, up to 48 characters.
- The finished file is one vertical MP4, 1080 by 1920, with the sound levelled. You download it and post it yourself.
- The export costs 30 credits per started minute of the finished video.
Questions people ask
- What is the best way to edit talking head videos?
- Cut first, then add captions, then show what you talk about, then add music and an ending. In that order, you never have to move captions or screenshots again after a cut.
- How long does it take to edit a talking head video?
- By hand, the cutting takes longest, because you listen to every second; captions go quickly if CapCut writes them and you only fix the mistakes. In Cutroom's Magic Editor you press "Make my video" once. It normally takes 10 to 15 minutes, and you get an email when your first cut is ready.
- What should I cut from a talking head video?
- Ums, uhs, false starts, repeated lines and long pauses. Keep a short breath between sentences, or the video sounds rushed.
- Can I edit a talking head video on my phone?
- Yes. The CapCut app can write your captions on your phone, and Cutroom's Magic Editor works in a phone browser, with the chat shown first.
- Can AI edit a talking head video for me?
- Yes. Cutroom's Magic Editor turns one clip of you talking into a finished vertical video: Jacob, its AI editor, makes the whole edit, and you change it by asking in a chat. Your first video is free.
- Can I edit a long talking head video the same way?
- Yes, the four steps are the same for a long video. You'll need a timeline editor such as CapCut or Descript for it, though. The Magic Editor takes a clip of up to 3 minutes and makes one vertical video.
Film one clip of you talking, press "Make my video", and Jacob does the four steps for you. Your first video is free.