Turn two photos into a personal elevator video inspired by (500) Days of Summer. Upload one portrait for each person to create the headphone conversation with you and a friend.
15 seconds · 9:16 vertical HD · no prompt needed.
Full 14 s0:00
Sample · 14 s · 16:9
AI-generated example · created with two sample portraits · enable sound to watch the full scene
What you get
Landscape sample16:9 · 14 s
Full scene14 seconds
Sound onReference audio
The public sample clips use the 16:9 landscape format. Choose Vertical for a 15-second 9:16 MP4 built for TikTok, Reels and Shorts; both exports are watermark-free.
TWO PORTRAITS · ONE ELEVATOR
Your elevator moment, in two photos.
Turn two photos into a personal elevator video. Upload one portrait for each person to create the headphone conversation with you and a friend.
The left slot listens through headphones. The right slot starts the exchange.
15sVertical elevator scene
2Separate portraits
9:16HD MP4 · no watermark
∞Download anytime
WATCH BEFORE YOU CREATE
Watch the full scene before you create.
The exchange is short and quiet: one person listens through headphones, the other recognizes the music, and the listener turns. Play the public examples with sound and watch the whole clip, not just the opening frame.
Your two portraits guide the people and their visible clothing. The template supplies the lift, the headphone gesture, the speaking turns and the reference sound. Upload order fixes the two roles.
Each slot takes one JPG or PNG portrait of up to 10 MB, checked in the browser before a local preview appears. One generation uses 150 credits and delivers a 15-second 9:16 vertical HD MP4 by default; a 14-second 16:9 landscape option is also available. Completed work is tied to your signed-in account with no download deadline. The three public clips below are labeled 16:9 examples, not renders of visitor photos. AI resemblance can shift during turns, so review the full result.
Last reviewed November 2026. This independent AI reinterpretation is not affiliated with the film, cast, band or social platforms. Get permission for the people you upload and check your audio rights before posting.
THREE LABELED PUBLIC EXAMPLES
See the scene in motion.
Three labeled I Love The Smiths AI examples made with two sample portraits. Enable sound to watch the full scene.
AI-GENERATED EXAMPLE
01
Selected test sample
Watch the headphone movement, the speaking turns and the reactions all the way through.
Read the scene description
Two consenting adults stand together in a softly lit lift. The left person wears headphones; the right person notices the music and speaks. The listener takes off the headphones and turns. Both stay visible in the 14-second clip.
AI-GENERATED EXAMPLE
02
Sample A
Check the head turns and both reactions before you create your own version.
Read the scene description
A second labeled example of the same fixed performance. Both cast members keep their sides through the exchange. This public sample does not predict how a future task will handle a specific pair of images.
AI-GENERATED EXAMPLE
03
Sample B
A landscape frame keeps the listener and speaker visible together.
Read the scene description
A third labeled example with both people in one wide frame. Review identity, wardrobe and timing before sharing, and check the audio rights on your destination platform.
Your portraits guide the two cast members while the template supplies the lift, the gesture, the speaking turns and the sound. No filming together and no editing timeline. The trend guide covers how people post the result.
The conversation unfolds in three small beats
01
A shared song starts it
The right-hand character makes the first move. The listener remains on the left, headphones on, until the remark catches their attention.
02
The listener leans into the moment
Removing the headphones makes the reaction visible. This gesture and the turn toward the speaker are part of the guided performance.
03
Recognition changes the mood
A glance and a restrained smile finish the idea. Each character follows the speaking turn assigned to that side of the source scene.
FROM TWO PHOTOS TO YOUR OWN ELEVATOR VIDEO
Four steps to your scene
Video tasks require credits. One-time packs are available on the pricing page.
01
Give each person a role
Two separate JPG or PNG portraits, up to 10 MB each. Pick sharp, well-lit faces. The left slot listens; the right slot starts the exchange.
JPG · PNG · 10 MB max each02
Sign in and choose a pack
Sign in, then choose 1, 3 or 10 videos. Each uses 150 credits. One-time packs start at $7.90.
150 · 450 · 1500 credits03
Let the scene take shape
Generation usually takes 5–15 minutes; queues and retries can add time. Follow the existing task instead of submitting another one.
One task · automatic retries04
Play it, check it, save it
Review both people throughout the clip, then download the MP4 any time from your library. A completion email links back to the finished video.
Email-code and Google sign-in are available. Video tasks require credits, and one-time packs are available on the pricing page.
What does the generator change?
Your two portraits guide the people and their visible clothing. The template supplies the lift, the gesture, the speaking turns and the sound. Upload order fixes the two roles; review the full result because likeness can shift.
Do I need to film anything or write a prompt?
No. Upload one portrait per person. The scene is already directed; the optional appearance field adds a visual preference without changing the dialogue or roles.
How much is one video, and is this a subscription?
One video costs $7.90, three cost $19.90 and ten cost $54.90. The packs add 150, 450 and 1,500 credits. One generation uses 150 credits. These are one-time purchases, not recurring subscriptions.
What happens if a stage fails, and why can the wait exceed 15 minutes?
A confirmed failed stage is retried automatically up to twice, and a final failure returns its credits. The 5–15 minute window is an estimate: model queues and delivery processing add time. Do not start a second task while a retry runs.
Can I change the dialogue, voice or video format?
The performance and soundtrack are fixed, and the site does not record your voice. You can choose the output shape: a 15-second 9:16 vertical HD MP4 by default, or a 14-second 16:9 landscape MP4. See the specifications above.
How long can I download my video?
There is no download deadline. Finished videos stay in your account, and the MP4 can be downloaded from My Videos whenever you need it. Keep your own copy before posting.
Does the export have a watermark?
No. Both the 9:16 vertical and 16:9 landscape exports are delivered as clean MP4 files without a watermark, ready for TikTok, Reels and Shorts.
Is this an official movie or band product?
No. This is an independent AI reinterpretation, not affiliated with the film, cast, band or social platforms. For the origin of the phrase, see our meme explainer. Get permission for both people and check audio rights before posting.
Which photo formats and sizes are accepted?
Each slot accepts one JPG or PNG image of up to 10 MB. The browser checks type, size and decoding before showing a local preview; MDN's File API reference describes that mechanism.
How do I buy credits?
Choose one of the three one-time packs on the pricing page and complete payment in Stripe Checkout. The balance is added after Stripe confirms the payment.
HD
CAST WITH PERMISSION
Use portraits you have the right to use.
Only use your own adult portrait or get permission from both people. Submitted portraits are sent to the generation service. Do not present generated work as a real event or endorsement.