Sump pits want Vidu water, not a voiceover
Vidu Q3 image-to-video puts diegetic pump sound on the pit still; a closed-mouth TTS read is a different ad.
Vidu Q3 image-to-video puts diegetic pump sound on the pit still; a closed-mouth TTS read is a different ad.
A sump listing is water in a basin you already photographed: the liner, the stand, the discharge, the lid gap. The buyer is matching a pit they have or a pit they are about to cut. The sound that belongs on that clip is the pump, the water, the check valve. A constructed voice walking "when you should install" is a different file. Do not glue it onto the water.
The pit still is the job
Photograph the basin you actually install or sell. Same lid, same crock, same discharge direction. Text-to-video will invent a cleaner basement with a cousin pump and a liner you do not stock. That is a failed SKU, not a cinematic upgrade.
AI image to video keeps frame one as the file you sent. Labels, floats, and the cracked concrete behind the basin survive because they were already in the picture. The prompt is for physics — water rising, a discharge pulse, a lid that does not grow a new pump. Re-describing the basement in the prompt is how the model starts drawing a second house.
If the still includes a homeowner's face, crop it. You do not need a person for this ad. You need the pit.
Vidu Q3 I2V is water plus diegetic pump
Vidu Q3 Image to Video is the still-to-motion row with native audio: 5s, 10s, or 15s, 1080p, 16 credits, discounted. Audio is on. Write the mix you want — a pump kick, water against liner, a quiet basin — or you will get a soundtrack you did not choose. A silent brief still gets sound. Prompt the water, not a lecture.
Iterate at 5s. Fifteen seconds of the same pit is a long miss unless the water actually travels. Aspects follow the still. Wide if the crock is the hero. Vertical if the feed is. The prompt will not mint a walk-out basement because you said "cinematic."
Diegetic means the sound is in the picture: impeller, water, pipe. It is not a host. It is not a hydrologist. It is not a "signs your basement is flooding" read. The catalog row is image-to-video with audio, not a presenter meter.
If you needed a last frame — dry pit to a pumping pit — that is a first-last job on a row that accepts two stills. This Vidu slug is one still, then motion. Photograph the state you want to start from.
A closed-mouth TTS read is a different ad
Gemini 3.1 Flash TTS is a 4-credit text-to-speech row: thirty voices, tags, multilingual synthesis. That is a voice file. It is not pit physics. Laying a closed-mouth read under the basin — "here is what you should do with this water" — turns a machinery clip into a talking explainer. The mouth does not even have to be in frame. The voice is the presenter.
Keep lips out. Keep advice out. If a real named installer must speak, that is footage of that person on a different day, or it is overlay type you wrote. Add a text overlay burns one authored line — a model, a GPH, a "lid on" CTA — whole clip, top / center / bottom. Vidu will not letter the discharge size. Overlay after the water locks.
Do not ask Vidu to typeset a flow rate on the lid. Do not ask TTS to prescribe a pit depth. The first is invented glyphs. The second is a lecture.
/for/plumbers is the crew; this is the pit
AI video for plumbers is the nearby audience: the satisfying fix, a clear pipe, a diagnostic walkthrough. It is written to a crew that films a job. It is not a sump-pit SKU hold. Do not rewrite that page as a basement hydrology talk, and do not paste its "signs of a hidden leak" voiceover idea onto this still.
This job is the basin. Installers who need crew proof can still film the real pull. Generation can still do the water on the photograph you already have. Mixing them in one generate is how a pit clip becomes a constructed adviser. Seasonal HVAC is a different still: furnace, not crock. Name the object.
FAQ
Can I prompt Vidu "no voiceover" and still get a host?
If you describe a person, you can still get a person. Hold the pit. Prompt water and pump. Overlay the line you wrote. Do not add a TTS bed "just for watch time."
Does Vidu Q3 image-to-video invent the pit if I only write a paragraph?
No — this slug requires the still. Frame one is the file. Text-to-video is a different row and will invent a basement. Upload the crock you actually have.
Why not Gemini TTS under a silent pit loop?
TTS is a voice file. Under a sump still it reads as a presenter. Diegetic pump on Vidu is the mix this ad wants. A real installer talking is a different shoot.
Is /for/plumbers the page for sump content?
It is the crew page: fix shots, diagnostics, van. This page is the pit still plus Vidu water. Do not rewrite the plumber audience as a catalog clip, and do not replace the basin with a talking explainer.