How to Add an AI Voiceover to CapCut (Without the Robot Voice)
Updated September 3, 2026
TL;DR: generate the voiceover in LeapFun (free to start, mastered MP3 out), import the file into CapCut, drop it on the timeline under your clips, trim to the cuts, export. Ten minutes end to end. Menu names below follow CapCut's current desktop layout and may differ slightly by version and platform.
Step by step
- Write the script and pick a voicePaste the script into LeapFun Text to Speech, choose a voice — narration, ads, social or educational — and tap a card to hear a real sample first.
- Generate and downloadGenerate, listen, re-take if needed, then download the MP3. Every export is loudness-mastered to -14 LUFS / 48 kHz, so it sits at platform loudness without a mixing pass.
- Import the audio into CapCutIn your CapCut project open the Audio tab and import the MP3 from your device (on mobile: Audio → Sounds → import from files). It lands in your project media.
- Add it to the timeline and alignTap the file to add it as an audio track under your clips. Drag to align the first word with your first cut, then split and shift segments to land on later cuts.
- Duck the music and exportLower any music track under the voice (or use the voice's volume control), add a short fade at each end, then export at your platform's resolution.
Start with a voice that fits the video — these are real LeapFun samples:
Tips that save a re-export
- Generate lines separately for fast-cut videos. For a montage, one file per sentence lets you snap each line to a cut instead of nudging one long track.
- Don't let CapCut's auto-captions fight the script. Use auto-captions for timing, then paste the exact script text to fix any misheard words — or generate captions from your script directly.
- Leave the voice at its exported level. It's already mastered to -14 LUFS; pull the music down rather than pushing the voice up and clipping.
- Clone your own voice for creator-style videos. If the channel is you on camera, a LeapFun clone (Starter plan) keeps the voice consistent across CapCut projects.
Frequently asked questions
Why not just use CapCut's built-in text-to-speech?
For quick trend videos it's fine and free. For anything where the voice carries the video — explainers, ads, storytelling — a natural voice with proper mastering and the ability to re-take one line is worth the extra step. Both approaches import the same way.
Can I use LeapFun audio in a monetised or sponsored CapCut video?
Yes on a paid LeapFun plan, which includes commercial usage rights. The free tier is for personal and evaluation use.
The voice and captions don't match — what happened?
CapCut auto-captions transcribe what it hears and can miss words. Correct them against your script, or generate captions from the script text rather than from the audio.
Does this work on the CapCut mobile app?
Yes. The import path is Audio → Sounds → import from your files; the alignment and volume steps are the same, with touch controls.