How to make a voiceover from text for a video
No microphone, no recording session: write the script, generate the voice in Cutline and drop the result on the timeline next to your footage. Speech is synthesized on your device.
Create a voiceover in five steps
- Open the editor and add your video to the timeline.
- Open the Voice tab and add a voice. The voice is downloaded the first time and then runs on your device.
- Type or paste your script into the voiceover script field and start the generation.
- When it is done, the result appears as an audio clip on the Voiceover track. Drag it to line up with the picture and trim its ends if needed.
- Adjust the volume, add fades, and export to MP4 or WebM with
⌘E(Ctrl+E).
Timing the voiceover to the picture
- Write short sentences. Each one is easier to move and re-time than a long paragraph.
- Split the voiceover clip with
Sat a sentence break and slide the parts to match what is on screen. - If the voice runs long, trim the pauses or cut the script, then generate again.
- Lower the volume of any music underneath so the speech stays clear.
If generation is slow or fails
Cutline renders speech on your GPU when it can. If that fails, it tells you and retries on the CPU, which is slower but works. A long script takes longer, so splitting it into a few parts can make it easier to review.
Add subtitles to the voiceover
Once the voiceover is on the timeline, you can run speech recognition on it to create matching subtitles. The steps are in how to add subtitles automatically.
Frequently asked questions
Do I need an account or an API key?
No. The voice runs on your device, with no account.
Is my script uploaded?
No. Speech is synthesized in your browser.
Does the exported video get a watermark?
No. The exported file contains only your footage and your audio.