How to make an AI rap video from photos

You need a clear photo of each person, permission to use those photos and one idea for the rap. You do not need to write a whole song. The creator turns your chosen topic into an original short performance; the scene controls its visual direction, rather than selecting a fixed soundtrack.

This guide follows a birthday example: two friends, photographed separately, making a short rap to celebrate Maya's birthday. The example topics below are writing suggestions, not transcripts from generated videos.

1. Decide between one shared photo and two portraits

Use Together in one photo when you already have a picture with two clear faces. Use Two separate photos when you have a good portrait of each person, even if you were photographed in different places. For the birthday example, separate portraits let you use a recent photo of Maya and a recent photo of her friend without arranging a new photoshoot.

Choose JPG, PNG or WebP files up to 10 MB each. If a phone photo will not upload, check its format and export a supported copy. Keep a natural view of the face: sunglasses, a hand over the mouth or a strong beauty filter removes information the generation needs.

Before continuing, look at the photos at the size they appear in the upload slots. Can you distinguish both people without zooming? Are their eyes and mouths visible? Clear photos can help resemblance, but they cannot guarantee an exact likeness.

2. Write one occasion and one personal detail

A useful topic gives the verse something to say. “Birthday” names an occasion, but “Maya's birthday, celebrating the friend who turns every road trip into a singalong” also gives the song a detail to build around. Include only information you are comfortable submitting for AI processing.

Keep the request focused. A list of names, several life events and a paragraph of exact lyrics is too much to fit naturally into a short clip. If pronunciation of a name is essential, remember that the result still needs listening and review; typing the name does not guarantee the delivery.

The current creator requests English rap vocals. The topic is a creative brief, not an exact-lyrics editor, a recording of your voices or a request to reproduce a famous song. Leave it blank if you prefer the default upbeat theme about two friends celebrating a good day together.

3. Match the scene to the feeling of the rap

For the birthday example, an orange booth keeps the idea close to a shared-microphone performance. A luxury lobby changes the mood toward a dressed-up celebration; a street cypher points toward a more casual group-performance setting. The choice changes the direction you ask for, not the identity of the people in the photos.

The creator offers two orange-booth options, a studio booth, a luxury lobby, a concert stage and a street cypher. Browse the reference demos with sound to decide which direction fits. Those clips illustrate styles and are identified as third-party references; they are not examples of what your particular photos will produce.

4. Set the format, length and credit budget together

Start with the destination. A vertical 9:16 composition fits a phone-first post. A square 1:1 frame suits a square layout; 16:9 gives you a landscape composition. Choose before generation so that you do not have to crop one person out later. Changing the aspect ratio does not add credits.

For a short birthday message, the default is a useful place to begin: 10 seconds at 720p on Standard uses 10 credits. A longer duration allows more room for the performance, but costs more and is not a guarantee of a more coherent story. Higher resolution changes the requested image detail; it does not correct an unclear input photo or ensure better pronunciation.

Open Model & length, select the available options, check the total and choose Apply settings. Read the model and credit guide if you are deciding between Standard, Fast, Pro and Ultra. Buy a credit pack or subscription only after understanding the cost of the video you want.

5. Start only when the inputs look right

Check the two people, scene, topic and credit total once more. Confirm that you have permission to use the photos, then click Generate when you are ready. Picking a topic or changing a setting does not start a video on its own.

Photos remain in your browser until you start generation, when they are submitted for AI processing. Avoid refreshing or repeatedly submitting while you are waiting. Processing time varies, so follow the saved task status instead of assuming a fixed finish time.

6. Watch the complete video before posting

Do one pass with sound and another looking closely at both faces. A strong opening frame can hide a later face change or an awkward line. For Maya's birthday clip, check that the result keeps both friends recognizable and that the verse is appropriate for the person receiving it.

• People: are there two distinct performers throughout, without an unexpected extra person?
• Sound: can you hear the vocals and beat, and is the spoken content suitable?
• Timing: do the mouth movements and gestures look acceptable during the performance?
• Framing: are both faces comfortably visible in the chosen shape?
• Sharing: are you comfortable with the names and personal details included?

Download the finished video and replay the saved file before uploading it elsewhere. If you add a title or captions in an editor, keep them away from faces and check that the exported sound still plays. Follow the destination platform's rules for AI-generated media and any music you add.

What to do when something is wrong

For help, use the feedback icon or email support@airapvideo.studio. Include the task reference and what you expected to happen. You do not need to send more personal photos to report a status or credit problem.