How to Make a Talking Head Video for Free (Step by Step)

A talking head video is a video of one person speaking straight to the camera, framed from the chest up, with little else on screen. It is the format behind course lessons, product updates, sales intros and most YouTube explainers. You can make one for free: film yourself on a phone or webcam near a window, read from a teleprompter or short notes at lens height, then cut the mistakes in a built-in editor such as Clipchamp or iMovie. Switch to an AI talking avatar only when you need many versions, several languages, or no camera time at all.

What a talking head video needs to work

Three things decide whether people keep watching: your eyes, your voice and your pace. The viewer should feel you are talking to them, so the lens sits at eye level and you look into it, not at your own face on the screen. Your voice has to be clean, because people forgive soft focus long before they forgive echo or hiss. And the script must be tight. One idea per video is the rule that saves most first attempts.

Everything else is optional. You don't need a studio, a second camera or b-roll for a good result. A plain wall, a tidy shelf or a slightly blurred room works as a background. The goal is a frame that doesn't compete with your face. Keep it simple and spend your effort on the take itself.

Set up the shot: camera, light and sound

Start with what you already own. A recent phone camera usually beats a laptop webcam, so use the rear camera if you can see a script some other way, or the front camera if you need to watch yourself. Put the device on a stack of books or a simple stand so the lens is level with your eyes. Leave a little space above your head and put your eyes near the top third of the frame.

Light comes next. Face a window during the day so the light falls softly on your face; never sit with the window behind you, or you turn into a silhouette. If the light comes from one side, a white sheet or a piece of foam board on the other side fills the shadow. Turn off overhead lights that put dark circles under your eyes.

Sound matters most. Record in the smallest, softest room you have, since rugs, curtains and a sofa absorb echo. Get the microphone close: a wired headset or a clip-on lapel mic a hand's width below your chin will beat the built-in mic across the room. Do a ten-second test, listen on headphones, and fix the hum before you record anything long.

Write it short and read it naturally

Write the script the way you speak. Short sentences, plain words, one point per paragraph. Time yourself reading the script aloud; for a 90-second video, keep the words to what you can say in 90 seconds without rushing, usually a little over 200. Open with the payoff in the first line, give the steps or the reason, and end with one clear next action. Read it aloud once before you film and cut every line you stumble on.

You have two ways to deliver it. A teleprompter shows the text near the lens so your eyes stay close to the camera; Clipchamp's webcam recorder has one built into the work and school version, turned on from the three dots menu ^[1]. The other way is bullet notes taped right under the lens, which sounds more natural if you know the topic well. Either way, record in short chunks of a paragraph or two. Pause, then start the next section cleanly, so each cut lands on silence.

How to make a talking head video in Clipchamp

Clipchamp runs in Edge or Chrome and as a Windows app, and it records, captions and exports in one place. These steps follow Microsoft's own help pages.

1. Open a project, select the record & create tab and choose the camera option.
2. Allow access to your camera and microphone when the popup appears. Pick an external camera or mic from the drop-down if you have one.
3. Click the red record button. A three-second countdown runs, then recording starts. Click stop when you finish the section.
4. Watch the take back. Choose retake recording to redo it, or save and edit to send it to the timeline.
5. Click the captions icon in the preview, choose the language and click transcribe media. Fix any misspelled words in the transcript.
6. Watch the take on the timeline. On a work or school account, highlight the dead air or a stumble in the transcript tab, right-click and select delete; the clip splits with no gap left in the timeline ^[2].
7. Export the finished video as an MP4.

Three limits shape your plan. Every webcam recording is capped at 30 minutes, and resolution tops out at 1080p, depending on your camera. Captions work in one language per video, and you can download them as an .SRT file. On work or school accounts trimming runs through the transcript, so caption first and cut second.

A mini example

Say you run a small bookkeeping practice and want a 90-second welcome video for new clients. You write 220 words in three blocks: who you are, what happens in the first week, how to reach you. You record each block as a separate take, keep the best of each, add captions, cut the stumbles from the transcript, and export. That's about an hour, most of it spent on the script.

Cutting a phone recording in iMovie on iPhone

If you filmed on an iPhone, you can finish there. Open a movie project in iMovie and add your clips. To shorten a clip, tap it in the timeline and drag either yellow trim handle. To remove a stumble in the middle, scroll so the playhead sits just before it, tap the clip, tap the Actions button, then tap Split. Split again after the mistake, tap the bad piece and tap Delete ^[3]. iMovie puts a plain cut between the new clips, which is exactly what a talking head edit needs.

A quick swipe down over the playhead also splits a selected clip. That makes the edit fast once you get the rhythm: play, stop on the mistake, split, split, delete. A jump cut between two sentences looks normal in this format, so don't try to hide it.

When an AI talking avatar beats the camera

The free method breaks at a specific point, and it isn't video quality. It breaks on repetition. Every change to the script means a new take, new light and a new edit. That's fine for one welcome video. It stops working when a sales team wants 20 versions with different names, when the same training must ship in Spanish and French, or when a policy update has to go out every month and nobody wants to sit in front of a camera again.

That's where an AI talking avatar takes over. You record yourself once to create a digital version, or pick a stock presenter, then type the script and the tool renders the person speaking it. Change a line and you re-render instead of re-shooting. Translation tools can then produce the same video in other languages with a cloned voice. The trade-off is consent. Some avatar types ask the real person to record a consent step first, but not all do, so getting approval for any face and voice you use is on you.

Three products in our catalog handle this job in different ways, starting with HeyGen's AI avatars.

If you'd rather compare every option for generated presenters and scenes side by side, our guide to AI video generator tools groups them by situation. For the bigger picture of what AI video can and can't do yet, start with the overview of AI video generation.

The bottom line

To make a talking head video for free, film yourself at eye level facing a window, keep the microphone close, read a short script from a teleprompter or notes near the lens, and record it in chunks. Trim and caption it in Clipchamp, or split and delete mistakes in iMovie on iPhone. That covers one-off videos well. Once you need the same message in many versions or languages, or nobody can spare camera time, an AI avatar tool saves the re-shoots. Just get consent for any face and voice you use.