How to Sync AI Voiceover to Video

To sync AI voiceover to video, import the numbered WAV files from your abcVoiceover download, place them on an audio track in order, and move each line to its moment in the picture. Because every line is its own file, you can nudge, trim or swap one line without touching the rest. Script.txt then gives you a ready-made transcript for captions.

Decide what leads: voice or picture

Voice first for explainers, tutorials, documentaries and animatics: lay the narration down, then cut or animate the picture to it.
Picture first for edits that already exist, such as a product demo recording or footage you're narrating over: place each line where its shot is, and rewrite lines that don't fit.

Either way, it helps to write the script one line per shot or beat, so each file matches a piece of picture.

Getting the files onto the timeline

Unzip the download and import the script's folder into your editor's media bin. The files sort in script order by name: 001 Narrator.wav, 002 Narrator.wav and so on.
Select them all in name order and drag them onto an audio track. In most editors, including Premiere Pro, DaVinci Resolve, Final Cut Pro and CapCut, multiple clips dragged together are placed one after another.
Space them out. Slide each line to where it belongs in the picture, leaving natural pauses between them.
Put each voice on its own track for dialogue, so levels and effects can be set per character.

Cutting to the voice

Cut on the pauses between lines, not in the middle of a word.
Give the viewer a beat after an important line before the next one starts.
If a line is too long for its shot, don't speed it up much in the editor. Shorten the line in abcVoiceover and generate it again; it takes seconds.
For tight timing, see pacing and pauses.

Swapping a re-take

When you re-take a line, download again and find the file with the same number. Most editors can replace a clip in place: Premiere Pro's Replace Footage, or relinking the media in DaVinci Resolve or Final Cut Pro. Or simply drop the new file on the timeline where the old one was. File numbers stay the same as long as you haven't added, removed or moved lines earlier in the script, so add late additions at the end if the edit is already placed.

Captions and subtitles

Script.txt in each script's folder has every line in order, which is a ready-made transcript.
YouTube can time a transcript for you: in YouTube Studio's subtitles, upload or paste the text and use Auto-sync.
Editors with speech-to-text captions will transcribe the voiceover; correct them against the script.
Fix respellings. If you wrote a name phonetically to get it pronounced right, change it back in the captions.

Mixing

Voice should sit clearly above music and effects. Duck the music under speech, and set the final loudness for the platform: YouTube plays videos back at about -14 LUFS, so mixing louder gains nothing. See mixing with music and loudness standards.

Disclose it

YouTube asks creators to disclose realistic altered or synthetic content, which can include a cloned voice of a real person, and other platforms have similar rules. See how to disclose AI voices.

How to sync AI voiceover to video