How to narrate a food video to what the hands are doing
Upload the footage and tell Vizard Agent what the dish is. Vizard Agent locates every visual beat in the process, such as the tin opening, the first placement and the zest going on. It verifies each one is actually on screen, auditions voices, and writes narration that lands on the moment it describes.
What do you type into Vizard Agent?
Describe the dish and the voice, not the edit. Vizard Agent works out the beats from the footage, so what it needs from you is the ingredient names and a clear picture of who is meant to be speaking over it. Name the ingredients exactly: they're spoken aloud, and a wrong one is worse than none.
Prompt
What does Vizard Agent actually do?
Food narration fails when the words and the hands are out of step. A line about lemon zest that plays while someone is still arranging fish reads as generic marketing, so the beats are found in the footage first and the lines are written to them. In the session behind this article it:
- Built a contact sheet of a single take, located the key visual beats and measured the source audio's loudness.
- Pulled a frame grid for each beat, from the tin opening and the toast to the first placement, and reviewed every grid by eye, including the onion and the zest.
- Zoomed into the sprinkle moment to confirm the lemon zest was actually visible before writing a line about it, and checked the hero frame at full size.
- Searched the voice library, generated voice tests and listened to them before choosing one.
- Generated the narration, pinpointed the placement beats again and assembled the track against them, for a 39.83-second vertical Reel.
When does this not work well?
Vizard Agent can only narrate what's genuinely on screen, so a beat that wasn't filmed can't be narrated: no zest shot, no zest line. One fixed wide angle leaves no close-ups to cut to, and a rushed build gives each line under two seconds. A generated accent suggests a region; it doesn't come from one.
Common questions
Will it invent a step that is not on screen? No. Vizard Agent verifies a moment is visible before narrating it, zooming in when unsure.
Can I hear the voice before committing? Yes. Vizard Agent generates voice tests and listens through them before generating the narration.
How long should a food reel be? Thirty to forty seconds. That is six or seven beats with room to say something about each.