Listing photos · slow camera moves · voice-over · music
AI Real Estate Video Generator — listing photos into walkthrough clips, honestly
Turn each listing photo into a short clip with a slow, steady camera move, add a voice-over and music, and cut them into a walkthrough-style video — without adding a single feature the home doesn't have.
- One clip per room photo
- 16:9 for tours · 9:16 for reels
- Voice-over and music
- No watermark
Updated by the Kenerate AI team
On this page
Quick answer
How do you make a real estate video with AI?
An AI real estate video generator turns still listing photos into short moving clips — a slow push-in on the living room, a gentle pan across the kitchen — and strings them into a walkthrough-style video with music and a voice-over. On Kenerate AI each photo becomes a clip in Video Studio; Wan 3.0 costs 15 credits a second at 480p, so eight 5-second room clips at 720p come to 1,200 credits. The model invents whatever the photo doesn't show, so check every clip for added windows, doors or views, and label the video as AI-enhanced wherever your MLS requires it.
Key facts
- How it works
- One listing photo per clip → Video Studio, Image to Video → video editor
- Clip length
- 2–30 s (Wan 3.0) · 3–15 s (Kling v3) · 4–15 s (Seedance 2.0)
- Ratios
- 16:9 · 9:16 · 1:1 · 4:3 · 3:4 (Wan 3.0)
How to
How to make an AI real estate video from listing photos
- Time needed
- Steps
- 4 steps
Step 1
Fix the photos first — honestly
Brighten, straighten and colour-correct in the AI real estate photo editing tools. Don't remove power lines, change the lawn or swap the view. Empty rooms can be virtually staged, labelled as “virtually staged”.
Step 2
Put the tour in order
Front of the home, entry, living room, kitchen, dining, bedrooms, bathrooms, outdoor space. Eight to ten of your best photos make a 40–50 second video.
Step 3
Animate one photo at a time
Open Video Studio → Image to Video, choose Wan 3.0, add the photo as the start frame and ask for one slow move: “Slow, steady push-in toward the fireplace. Nothing in the room changes; no new objects, windows or doors.” Draft at 480p (75 credits for 5 seconds).
Step 4
Check every clip frame by frame
Look for anything the room doesn't have: an extra window, a doorway, a longer counter, a different view, furniture that wasn't there, bent walls. Re-run with a gentler move or drop the clip.
See it in the app

Open Video Studio in Text to Video mode and write the shot — subject, action, camera move and light.
Watch the tour · 1:19
Text, image & reference to video, models, edit, extend & chat
Chapters
Checked 22 October 2026
Which real estate video do you need?
Three kinds of listing video, and how each is made on Kenerate.
| Single room clip | Listing walkthrough | Social reel | |
|---|---|---|---|
| Photos | 1 | 8–10, in tour order | 4–6 of the best |
| Length | 4–5 s | 40–50 s | 15–30 s |
| Ratio | Same as the photo | 16:9 | 9:16 (crops the sides of landscape photos) |
| Sound | None or room tone | Voice-over + music | Music + captions |
| Made in | Video Studio | Video Studio → Voice Studio → video editor | Video Studio → video editor |
| Example cost | 5 s Wan 3.0 at 720p: 150 credits | 8 × 5 s at 720p + voice + music: ≈ 1,266 credits | 6 × 4 s Wan 3.0 at 480p: 360 credits |
Sources: NAR — 2026 Code of Ethics and Standards of Practice (Article 12) · CRMLS — Digitally altered images FAQs · NorthstarMLS — Guidelines for virtual staging and AI-enhanced listing photos · Florida Realtors — Your AI listing video needs guardrails (Aug 2026) · 24 CFR 100.75 — Discriminatory advertisements, statements and notices
Engines
The models behind a listing video
From the app's own model lists, 22 October 2026. Credits are per second of video, per character of speech or per track.
- Default
Alibaba
Wan 3.0
- Input
- Listing photo + camera-move prompt
- Output
- 2–30 s · 480p–1080p · 15 credits/s at 480p, 30 at 720p
Open Wan 3.0Best for
Room-by-room clips at the lowest cost; keeps a ratio picker
Kuaishou
Kling v3 Standard
- Input
- Listing photo + prompt
- Output
- 3–15 s · 16:9, 9:16, 1:1 · 25.2 credits/s silent
Open Kling v3 StandardBest for
Silent clips when the voice-over carries the sound

ByteDance
Seedance 2.0
- Input
- Listing photo + prompt
- Output
- 4–15 s · 480p–4K · 72 credits/s at 720p
Open Seedance 2.0Best for
Hero exterior or view shots at higher resolution

Google
Veo 3.1 Fast
- Input
- Listing photo + prompt
- Output
- 4, 6 or 8 s · 16:9 or 9:16 · 30 credits/s silent
Open Veo 3.1 FastBest for
A polished opening shot of the front of the home
- Voice

ElevenLabs
ElevenLabs Eleven v3
- Input
- Your listing script
- Output
- Speech · 60 credits per 1,000 characters
Open ElevenLabs Eleven v3Best for
The voice-over, in 70 languages
- Music

Kenerate
Ken Music 1.5
- Input
- A music prompt, instrumental
- Output
- Up to ~3 min · 30 credits a track
Open Ken Music 1.5Best for
A calm music bed under the tour
Video Studio needs a paid plan. Voice Studio and Music Studio bill per character and per track.
Tips
Listing clips that stay true to the house
Which moves are safe, and where AI starts inventing rooms.
Push-ins are safest: moving toward what's already in the photo reveals nothing new for the model to make up.
Pans, orbits and pull-outs show space beyond the photo's edges — exactly where fake windows, doors and furniture appear.
Keep clips at 4–5 seconds. Longer clips drift further from the photo.
Use the end-frame slot only with a second real photo of the same room; the model invents everything between the two angles.
Leave people and pets out of the prompt — they draw attention and aren't in the listing.
Keep the original photos next to the video in the listing, so buyers can compare.
FAQ
AI Real Estate Video Generator questions
10 questions
How do I turn listing photos into a video with AI?
Fix the photos honestly first, then animate them one at a time: in Video Studio choose Image to Video and Wan 3.0, add a photo as the start frame and ask for one slow camera move with nothing in the room changing. Check each clip, voice a short script in Voice Studio, add an instrumental music track, and assemble everything on the video editor's timeline.
Link to this answerHow much does an AI real estate video cost?
A 5-second room clip with Wan 3.0 is 75 credits at 480p or 150 at 720p. An eight-room, 40-second walkthrough at 720p is about 1,266 credits: 1,200 for the clips, about 36 for a 600-character ElevenLabs voice-over and 30 for a Ken Music 1.5 track. Video needs a paid plan; one-time packs start at $15 for 1,300 credits.
Link to this answerIs an AI listing video allowed on the MLS?
Usually, if it shows the property truthfully and is labelled where required — but rules differ by MLS and state, so check yours. NAR's Code of Ethics requires a true picture in advertising, CRMLS requires altered images to be labelled with the originals available, and NorthstarMLS's proposed guidelines bar edits that change walls, floors, windows, doors or other permanent features. An AI clip made from a photo is altered media.
Link to this answerDo I have to disclose that the video is AI-generated?
Often yes, and it's good practice anyway. California's AB 723 and MLS rules such as CRMLS's require labels like “digitally altered” or “virtually staged” on altered listing images. Put a short “AI-enhanced from listing photos” caption in the video and keep the original photos in the listing.
Link to this answerCan the AI add features the house doesn't have?
It can, by accident — and you must not publish them. When a camera move shows space outside the photo, the model fills it in and may add a window, a doorway, a nicer view or furniture. Use slow push-ins, keep clips short, check every frame and drop any clip that changes the property.
Link to this answerWhich camera moves work best for real estate?
A slow push-in toward the room's focal point (a fireplace, a window, the kitchen island) is the safest, because it reveals nothing the photo doesn't already show. A slight upward tilt works for exteriors. Avoid fast orbits, fly-throughs and pull-outs, which make the model invent space.
Link to this answerCan I make a vertical reel for Instagram or TikTok?
Yes. Wan 3.0 keeps a ratio picker in image to video, so you can choose 9:16 — but it crops a landscape photo, so frame the subject in the centre. The video editor also has a 9:16 project setting for the final cut, with a text track for captions.
Link to this answerCan AI do the voice-over too?
Yes. Paste the script into Voice Studio and pick a voice; ElevenLabs Eleven v3 costs 60 credits per 1,000 characters and reads 70 languages. Write updates as years (“new roof in 2023”) rather than ages — the content filter blocks phrases like “5 years old” — and describe the home, not the kind of buyer it suits.
Link to this answerCan I use AI to virtually stage the rooms in the video?
Stage the still photo first with the virtual staging tool, label it “virtually staged”, then animate it. Keep the unstaged original in the listing. Don't stage over permanent features or change finishes, and disclose staging in the video as well.
Link to this answerIs there a free AI real estate video generator?
Not on Kenerate: video needs a paid plan, with one-time packs from $15 and no subscription. New accounts get 25 credits for images, so you can try photo fixes before you buy a pack for the clips.
Link to this answerIn depth
AI Real Estate Video Generator, explained
On this page
Is an AI listing video allowed — and what do the rules say?
There's no single national rule for AI listing video, but the existing rules for listing photos already cover it, because a clip animated from a photo is altered media. Not legal advice — read your own MLS rulebook and state law:
- NAR's Code of Ethics, Article 12: advertising must present a true picture; Standard of Practice 12-10 bars misleading images [1].
- CRMLS: digitally altered images need a label — “digitally enhanced”, “digitally altered” or “virtually staged” — with the original available; removing power lines or changing landscaping isn't allowed, and AI-generated landscaping images aren't permitted [2].
- NorthstarMLS's proposed guidelines (July 2026): no edits that change or hide walls, floor coverings, doors, windows, roofing, siding, ceilings, fireplaces, driveways or grading [3].
- Florida Realtors, reposting an Inman column (August 2026): disclose alterations as a routine part of marketing, don't show nonexistent features or changed layouts, and review AI material for accuracy [4].
Why AI video invents rooms — and which moves are safe
An image-to-video model only knows what's in the photo. When the camera moves toward the subject, everything it shows was already there. When the camera pans, orbits or pulls back, the model has to paint space it has never seen — and it paints something plausible: another window, a wider doorway, a deck, a better view. That's the moment a listing video turns into a misrepresentation.
So use push-ins and small tilts, keep clips to 4–5 seconds, describe every permanent feature as unchanged in the prompt, and check each clip frame by frame before it goes anywhere near the listing.
Fair housing and the voice-over
The federal fair-housing advertising rule covers words, pictures and spoken statements that show a preference for or against a protected class, such as race, religion, disability or familial status [5]. Describe the property — rooms, finishes, updates, location — not the buyer it suits. Kenerate's content filter also blocks scripts that mention children or an age in years, so write “updated in 2023” rather than “5 years old”.
What other AI real estate video tools offer
From each vendor's own page, read on 22 October 2026 [6][7]:
- AutoReel: upload or import listing photos, choose push-in, pull-out or orbit motion, music, captions and an AI voice-over; it says disclosure depends on the MLS, brokerage and jurisdiction. No prices or video lengths on the page.
- Luma: AI agents turn property photos, floor plans and descriptions into marketing videos; no lengths, camera controls, prices or disclosure guidance on the page.
- Kenerate: you pick the model and the move for each photo, see the credit price before each clip, and assemble the voice-over, music and captions yourself — slower than a template, but every frame is your call.
1. Living room push-in (Wan 3.0 · 5 s)
Slow, steady push-in toward the fireplace. The room stays exactly as in the photo — same furniture, walls, windows, floors and view. No new objects, no people, no lighting changes. Smooth gimbal movement.
2. Kitchen glide (Wan 3.0 · 5 s)
Very slow, short slide to the right along the kitchen island, keeping everything in the original frame. Cabinets, counters, appliances and windows stay exactly as they are. Nothing new enters the shot. Bright, natural light.
3. Front of the home (Veo 3.1 Fast · 6 s · silent)
Gentle tilt up from the front walkway to the house façade. The house, roof, windows, landscaping and sky stay exactly as in the photo. Leaves move slightly in a light breeze. No people, no cars, no new plants.
4. Bedroom reveal for a reel (Wan 3.0 · 4 s · 9:16, subject centred)
Slow push-in toward the bed and the window. The room's furniture, walls and the view through the window stay exactly as shown. Soft morning light, no changes to the room.
5. A 40-second voice-over script (Voice Studio, Eleven v3)
Welcome to 24 Maple Lane. A bright three-bedroom, two-bath home with oak floors throughout. The living room opens onto a kitchen with quartz counters and a large island, updated in 2024. Upstairs, the main bedroom has a walk-in closet and an en-suite bath. Out back, a fenced yard and a covered patio. Call us to book a viewing.
Explore
Every Kenerate video tool
Generators, tools, models and guides — all in one studio.
Make your listing video
Your photos, your script, a price before every clip. No watermark.
Tried it? Tell us how it went





