Two portraits · one AI video
Upload two photos. Turn them into a directed 15-second AI video.
Two portraits · one performance
Use clear photos with each person’s face and upper body visible. Person 1 and Person 2 keep separate roles throughout the video.
720P · 1 credit · no watermark
Multiple paid videos can generate at the same time · failed runs refunded
Real scene previews
These short examples show the two-person visual direction and scene treatment. Generated details can vary with the photos you choose.
How it works
Choose a clear photo for Person 1 and Person 2, with each face and upper body visible.
Tip: similar lighting and eye-level angles give both character references clearer visual information.
Raindance places both people into the directed 15-second template with separate character roles.
Tip: keep the result page open while the video is processing.
Preview and download the finished clip when processing is complete.
Tip: save the finished file promptly because provider download links can expire.
The format
The Raindance AI video generator is a focused two-photo workflow for making a directed 15-second visual. Instead of asking you to describe two characters, a lens, environments and movement in a long prompt, Vidoro starts with one recognizable format. The photos provide Person 1 and Person 2; the template supplies separate roles, cinematic atmosphere and the performance direction.
People are sharing this type of clip because it transforms two still character references into something that feels directed. The performers move through a continuous sequence with coordinated timing, natural weight shifts and cinematic scene changes. The result is short enough to understand immediately in a social feed.
The Raindance AI video generator creates only the visual. It does not include a song, vocals, speech or a hidden audio track. The music associated with the format is copyrighted, so Vidoro does not redistribute it or combine it with a paid generation. After downloading the silent MP4, use TikTok or Instagram's own Add sound control and select an authorized sound from that platform's library.
That separation is intentional. It keeps the generated asset useful as a clean visual and lets each publishing platform handle music rights, availability and regional restrictions through its licensed catalog. It also means you can reuse the same visual on Reels or Shorts and choose an appropriate sound separately on each service.
Identity quality begins with both inputs. The model must infer two faces from separate images while also creating frames that never existed in either source photo. Clear eyes, visible jawlines, even illumination and unobstructed hairstyles give it more consistent evidence. Tiny faces, extreme angles or strong beauty filters remove detail the animation stage needs.
Audio and copyright
Vidoro delivers a silent MP4. The publishing platform is the right place to add music it is authorized to offer.
Create a new post and choose the 15-second video you downloaded from the result page. Confirm that both people remain visible before moving to the audio step.
Keep the clip at its original length and avoid stretching it. The motion is timed as a complete 15-second performance, so trimming the beginning can make the entrance feel abrupt.
Search for Raindance inside the platform's licensed music library and select the sound available in your region. Preview the alignment, adjust the start point if needed, then publish from the platform.
Practical fixes
Replace that person's input with a larger, sharper front view. Remove sunglasses, masks and photos where hair or hands cover the eyes, nose or jawline. Avoid screenshots that have been repeatedly compressed. Each source photo should contain only its intended person, with enough room around the head and upper body for a natural composition.
Use a neutral source pose with shoulders visible and avoid an extreme close-up. Strong side profiles make the model invent more unseen facial structure during movement. If hands are already crossing the face in the source, choose another photo so the animation begins from clean anatomy and an unobstructed identity.
Check that both files are JPEG, PNG or WebP, are under 10 MB each and each contain one visible person. A provider timeout or safety rejection can also stop a task. The result page never submits an automatic paid retry. When a paid task fails, its held credit is returned; review the status, replace the affected photo if needed and start a new attempt yourself.
Ready to share
The finished clip is 15 seconds long, giving the two-person performance time to establish a mood without asking viewers to wait through a long introduction. Before publishing, preview the platform's crop and make sure both characters remain visible.
Upload the finished file, review the crop, use Add sound and select an authorized library result.
Keep both characters inside the visible frame, then choose music from Instagram's licensed catalog.
Upload the visual, check the mobile preview and choose suitable licensed audio through YouTube's creation tools.
Music availability can differ by country, account type and platform. If a particular sound is unavailable, do not download and re-upload an unauthorized copy. Choose another sound offered by the platform or publish the visual silently.
Focused by design
General video generators are useful when you want to invent a new world from text, but they ask you to make many decisions. You may need to specify the camera, lighting, wardrobe, movement, duration and aspect ratio, then rewrite the prompt when one detail drifts.
This Raindance AI video generator is for a narrower goal. Two portraits supply the character references, a fixed template supplies the direction, and the service handles the scene and motion. You do not need editing experience or prompt-writing knowledge.
The fixed format also makes attempts easier to compare. If one pair of source photos performs better than another, the environment and output length remain consistent. You can judge the two identities and motion rather than wondering whether a rewritten scene description changed the result.
Choose this tool when you want this specific short-form visual. Choose a general generator when you need an original location, dialogue, a custom camera plan or an entirely different duration. A focused workflow is simpler, but it is not meant to replace open-ended filmmaking.
FAQ
A paid 720P generation costs 1 credit, and a paid 1080P generation costs 1.6 credits. Both are 9:16 and watermark-free.
Use two sharp photos with even light and each person's face and upper body visible. Avoid strong side profiles, sunglasses, heavy shadows, motion blur, hands over the face and large obstructions.
No. The downloaded video is silent because the original music is protected by copyright. Upload the finished visual to TikTok or Instagram and add the authorized sound from that platform's music library.
It is a directed 15-second AI video made from two portraits. Person 1 and Person 2 keep separate roles while the fixed template supplies the performance, cinematic environment and scene changes without requiring a prompt.
Most generations finish within three minutes, although provider traffic and image complexity can make an individual task take longer.
Yes. Download the finished MP4, then use the crop and layout controls in the destination platform to prepare it for TikTok, Instagram Reels or YouTube Shorts.
Generative results vary with the sharpness, lighting, angle, face size, hairstyle visibility and obstructions in both photos. Clear eye-level references give the model stronger identity information.
A failed paid task returns its held credit automatically. The result page does not submit a second paid request, so you can review the error and retry manually with clearer photos.
One-time credits
One credit makes one 15-second 720P video. Raindance 1080P uses 1.6 credits. Credits stay valid for 12 months, and failed generations are refunded automatically.
One-time pack
$7.9
$7.90 per video
Valid for 12 months
One-time pack
$19.9
$6.63 per video
Valid for 12 months
Most popular
$44.9
$5.61 per video
Valid for 12 months
One-time pack
$99
$4.95 per video
Valid for 12 months
