How to Add a Professional AI Voiceover to Your YouTube Video No Recording Equipment Needed

Before generating a voiceover, you need a script. This is the text that will be read aloud. You can write it yourself, or use the AI script generator to create one from a topic. Either way, the script should read naturally when spoken — short sentences work better than long academic ones.

On the project page, the shot list header has a "Generate Narration" button. It opens the voice picker, where the voices are grouped by language.

[Screenshot: the shot list header with the Generate Narration button, above shots that already carry takes.]

Narration is recorded from the shot list, one take per shot.

Use the language dropdown to filter voices by language. There are voices in English, Simplified Chinese, Traditional Chinese, Japanese, Korean, French, German, Spanish, and Italian.

Each voice has a play button so you can hear a sample before selecting it. Pay attention to accent and tone — some voices sound more formal (better for educational or documentary content), others sound more casual and energetic (better for entertainment or motivation).

Click "Generate Narration". Each shot is recorded as its own take, one after another, at 1 credit per shot. You watch each take land in its row rather than waiting on a single long job — and if one shot fails, it costs that shot, not the run.

Every row gets its own player, so you can listen to any line on its own. Each take also reports its real duration — that is what sets how long the shot holds the screen, and what the subtitles are timed from.

[Screenshot: shot rows with individual audio players and per-take durations.]

One take per shot, each with its own player and measured duration.