How to Add a Voiceover to a Video

To add a voiceover to a video, write a script timed to the footage, record it in a quiet room (or generate it with an AI voice), place the audio on the editor’s timeline under the matching shots, lower any music so the voice stays clear, and export. Most editors, including iMovie and CapCut, also let you record straight onto the timeline.

Below are the three common ways to do it, the settings that make a voiceover sound clean, and how to fix timing when the voice and the picture drift apart.

The short answer

Write a script that fits the length of your video (about 150 words a minute), then either record it with your phone or a USB microphone in a quiet room, use your editor’s record-voiceover button, or generate it with an AI text-to-speech tool. Drop the audio on the timeline, trim it to the shots, keep music about 15 to 20 dB below the voice, add captions and export.

Key takeaways

• Script first: about 150 spoken words a minute is a reliable starting point.
• A quiet, soft-furnished room matters more than an expensive microphone.
• Editors like iMovie record straight onto the timeline; separate recording gives you more control.
• AI voices are fast for drafts, translations and frequent updates; your own voice builds more trust on hero videos.
• Keep music well below the voice, and add captions for viewers who watch on mute.

Disclosure: we make QuickVeed and show how to do this in it, along with other ways. Details about other products come from their public pages, checked in October 2026 and linked at the end.

Step 1: Write a script that fits the video

A voiceover that is too long forces you to speed-read; one that is too short leaves dead air. Time the footage first, then write to it. Most narrators speak at roughly 130 to 160 words a minute; 150 is a good planning number.

Leave a little room: plan for 90 percent of the time so the voice can breathe and the picture can make its point. Our words to minutes calculator checks a script’s spoken length before you record.

Write for the ear. Use short sentences, say numbers the way you’d speak them, and describe what the viewer is about to see a beat before it appears. Mark each paragraph with the timecode of the shot it belongs to; that makes the edit much faster.

Step 2: Record it yourself (phone or computer)

The room matters most

Record in a small room with soft things in it: a bedroom with curtains, a closet full of clothes, or under a duvet if you must. Hard, empty rooms add echo that no plug-in fully removes. Turn off fans and notifications, and put your phone in airplane mode if it isn’t the recorder.

Mic and distance

A phone held about a hand’s width from your mouth, slightly to the side so you don’t pop “p” sounds, already beats a laptop microphone across the desk. A USB microphone is the next step up. Keep the same distance for the whole take, so the volume stays even.

Recording settings

• Record in WAV or a high-quality format; save MP3 for the final export.
• Set the level so your loudest words peak well below the top of the meter. Clipped audio can’t be repaired.
• Record 5 seconds of silence at the start. Noise reduction tools use it to learn the room’s sound.
• Read each paragraph two or three times. Picking the best take later is faster than redoing a session.

Free tools such as Audacity handle recording, noise reduction and trimming on Windows, Mac and Linux. On a phone, the built-in voice recorder app is fine.

Step 3: Add the voiceover in your editor

Record straight onto the timeline

Many editors have a record-voiceover button. In iMovie on a Mac, Apple’s guide says to move the playhead to where the narration should start, click the Record Voiceover button below the viewer, use Voiceover Options to pick the input and set the volume (and mute the project while recording if you like), click Record, then click it again or press Space to stop. The clip lands under the shot that was at the playhead.

This is the quickest way for short videos because you see the picture while you speak. The downside is that each retake means re-recording live.

Import a recorded or generated file

1. Import the voiceover file (WAV, MP3 or M4A) into your project.
2. Drag it to an audio track under the video, starting where the first line belongs.
3. Play it back and cut the audio at each paragraph break with the blade or split tool.
4. Slide each piece so it starts on its shot, or trim the shots to fit the voice.
5. Remove breaths and long pauses where they slow things down; leave short ones so it sounds human.

If the voice and picture drift apart over a long video, don’t try to fix it in one move. Cut the audio at every sentence and line each sentence up with its shot. For the reverse problem, a video that runs short, slow the footage slightly with a tool like our video speed changer, or add a B-roll shot.

Step 4: Balance the sound and add captions

Most voiceovers sit on top of music or the video’s own sound. The voice must win. A common rule of thumb is to keep music about 15 to 20 dB below the voice while someone is speaking, and let it come back up in gaps. Many editors can do this automatically (often called ducking).

Then add captions. Many people watch with the sound off, and captions also catch any word the voice blurs. Generate them from the finished voiceover, check names and numbers, and either burn them in or upload an SRT file to the platform. Our guide to burning subtitles into a video covers both.

Common voiceover mistakes (and quick fixes)

• Reading, not talking. A read script sounds flat. Imagine one person across the table and speak to them; smile slightly on upbeat lines, it changes the tone.
• Describing what’s already on screen. “Here you can see a dashboard” wastes words. Say what it means: “Every unpaid invoice, in one list.”
• Voice and picture out of step. The viewer should hear about a thing just as, or just before, it appears. Nudge clips by a few frames until it feels natural.
• Different takes, different sound. Recording on two days in two rooms makes cuts audible. Record the whole script in one session, or re-record the short section that changed in the same spot with the same mic distance.
• No room for the visuals. A voice that never stops leaves no time for a product shot to land. Leave a one-second gap after key moments.
• Forgetting the end. Finish on the action you want, said plainly, and hold the last frame for a second after the voice ends.

Before you export, play the whole video once on a phone speaker and once on headphones. Phone speakers reveal a voice that is too quiet next to the music; headphones reveal clicks, breaths and background hum.

Step 5 (optional): Use an AI voice instead

AI text to speech makes sense for drafts, frequent updates (a price change shouldn’t need a studio session), translations, and people who don’t want to record. The workflow is the same as above, except the file comes from a tool instead of a microphone.

Doing it in QuickVeed

1. Open AI voice and paste your script, up to 10,000 characters.
2. Pick a model and a voice. QuickVeed has 9 text-to-speech models, from Chatterbox (7 credits per 1,000 characters) to ElevenLabs v3 or MiniMax Speech 2.8 HD (25 per 1,000).
3. Optionally clone your own voice from a short sample on MiniMax 2.8, Chatterbox, XTTS v2 or OpenVoice. Only clone a voice you have permission to use.
4. Generate, listen, and fix any line that sounds wrong by rewording it (for example, writing a number out in words).
5. Download the audio and drop it on your editor’s timeline as in Step 3.

Cost example: a 60-second script of about 150 words is roughly 850 characters, so one take costs about 25 credits on ElevenLabs v3 or about 7 on Chatterbox. QuickVeed doesn’t lay the voice onto your video for you; you place it in your editor. Then add captions in captions and subtitles (1.5 credits a minute for the transcript, 1 a minute to burn in one of 11 styles). If you want the video in another language with the original speaker’s voice, that is dubbing rather than a voiceover; see how to dub a video.

Frequently asked questions

How do I add a voiceover to a video on my phone?

Use an editor app with a voiceover or record button, such as iMovie on iPhone or CapCut. Place the playhead where narration should start, tap record, speak, and stop. Or record in a voice memo app and import the file.

How do I add a voiceover to a video for free?

Record with your phone or Audacity (free), then add the file in a free editor such as iMovie, CapCut or DaVinci Resolve. AI voice tools often have free tiers, but check whether they allow commercial use.

How many words do I need for a 1-minute voiceover?

About 140 to 150 words. Read it aloud with a timer, because pace varies by speaker and style.

Can I use an AI voice for a voiceover?

Yes. Generate the audio from your script with a text-to-speech tool, download it and place it on your timeline. Check the tool’s commercial terms and don’t clone anyone’s voice without consent.

Why does my voiceover sound echoey?

The room is too hard and empty. Record in a smaller room with curtains, a rug or clothes around you, and keep the microphone close.

Try it with your own video

Turn your script into a voiceover with QuickVeed’s AI voice and drop it on your timeline. Paste your script, pick a voice model, and judge the take for yourself. QuickVeed plans start at $18/month, and new accounts get a 14-day trial with 75 credits, no card needed. Sign up or see pricing.

Sources

Competitor details in this post were checked against these public pages in October 2026. Products and prices change, so confirm with each vendor before you decide.

• Apple iMovie User Guide: record a voiceover