Ferry ads lock the vessel still; water motion comes after the hull holds
The named boat is the SKU. Vidu for wake once identity holds. TTS the timetable as a voice file, not a talking captain.
The named boat is the SKU. Vidu for wake once identity holds. TTS the timetable as a voice file, not a talking captain.
A ferry ad that books a crossing is this hull: fleet number, paint, the bow a commuter already knows from the slip. Water is physics after that identity holds. A constructed captain walking "you should take the 7:10" is a different file.
The named hull is the plate
The honest still is a photograph of the vessel you operate — same livery, same name board. Text-to-video will invent a prettier ferry and seat a crew you did not cast. That is a failed SKU. If the still is already wrong, every second of wake is waste.
AI image to video keeps frame one as the file you sent. The prompt is for physics — a slow hold, light on the waterline, no new superstructure. Re-describing the boat is how the model draws a second hull. If the still includes a passenger's face, crop it. A generated commuter is a private extra. A generated captain is a replica.
Vidu for wake once identity holds
Wake is water: foam, hull-side chop, a horn you write into the mix. It is not a second boat. Once the named hull is locked, water is the later job. Do not ask a paragraph to invent both.
Vidu Q3 Video is Vidu's text-to-video engine. Audio is on. Durations 5s, 10s, 15s. 1080p. 16 credits. No image required. A silent prompt still gets a soundtrack. Write the water: wake against hull, light spray, no speech. A 5-second take is the shortest clip on this slug. Use T2V only when you accept generic water as atmosphere. The moment the brief is this boat, leave text-to-video.
When the hull must survive, start from the still. Vidu Q3 Image to Video is the still-to-motion sibling: image required, native audio, 5s / 10s / 15s, 1080p, 16 credits, discounted on that row. Prompt the water, not a second ferry. Iterate at 5s.
Do not put a sailing time in the Vidu prompt. Overlay the string from the timetable you actually run. Add a text overlay burns one authored line on the whole file. Type after the hull is a keeper.
TTS the timetable as a voice file
Gemini 3.1 Flash TTS is a 4-credit text-to-speech row: thirty voices, natural-language style control, inline tags ([sigh], [laughing], [whispering]), multilingual synthesis, multi-speaker dialogue. That is a voice file. It is not hull physics. It is not a talking captain.
Read the timetable you already posted. One crossing, one clock. Lay that file under the locked hull clip. Add voiceover to video is the finish: audio on a picture you already trust. The face stays off.
A closed-mouth TTS read under a generated captain is still a presenter. The voice is the host. "Here is what you should do if you miss the last boat" from a synthetic skipper is a persona plus a prescription. A real named captain is a different day and a grant. Generation can still do the hull. It cannot be the crew.
Write "no speech" on the Vidu take if the mix should stay water. Then decide whether the timetable is overlay type, a TTS bed, or both. Do not ask Vidu to typeset the 7:10 on the bow. Do not ask Gemini to invent a sailing operations did not publish.
/for/travel-agencies is the destination reel
Travel agencies own cinematic destination: a reel that is not this crossing. Do not rewrite it as a harbour film, and do not paste a "dream weekend" voiceover onto a commuter hull. A tourism paragraph — "luxury ferry at golden hour, happy families on deck" — restages the SKU and seats extras. Minors are a hard stop. Do not generate a crowd because a travel reel "needs life."
Hotels are further out: the property, the stay. A hotel that I2Vs your ferry jpeg as a waterfront transfer has used the wrong still. A ferry operator who copies the hotel playbook has used the wrong still the other way.
One named hull. Wake after identity. Timetable as a voice file or as overlay. Not a talking captain. If the name board grew a second serif, you used text-to-video on a SKU.
FAQ
Can I text-to-video the named boat if I describe the livery carefully?
You can. You will recast the number and the superstructure every take. Photograph the hull so those are not sampled again. Vidu Q3 Video is 5s / 10s / 15s, 1080p, 16 credits, audio on, no still required — which is how you get a harbour instead of a SKU. I2V the vessel still when identity is the job.
Why TTS instead of generating a talking captain?
Gemini 3.1 Flash TTS is a speech row: 4 credits, thirty engine voices, a timetable as a file. A talking captain is a video generate plus a likeness. Cast a real named person, or keep the face off the line. The hull still does not need a mouth.
Does Vidu Q3 Video require a still of our ferry?
No. That slug is text-to-video. Image-to-video is the sibling that requires the photograph. Use T2V for water atmosphere you will not claim as the named boat. Use the still when the fleet number has to hold.
Is /for/travel-agencies the page for ferry content?
It is the destination page. This page is the named vessel still, wake after identity, timetable as TTS or overlay. Do not rewrite the agency reel as a catalog clip, and do not replace the hull with a talking skipper.