Hair salon promo video with AI, from your own photos

A hair salon promo video is a short vertical clip of your real salon, your stylists at work and finished looks from real clients, with a spoken offer and music, made for Reels, TikTok or Shorts. With AI you can make one from photos you already have: animate each photo into a short clip, join the clips over a voice-over and a music bed, and render at 9:16. The same plan works for beauty and nail salons.

The Sume steps come from the Video generation, Timeline 1.0, Music Router and Video captions docs and the Sume API reference for text-to-speech, read on 2026-09-29.

What shots should a salon promo video include?

Four to six photos, each given one small movement:

• The space: the chairs, mirrors and wash station, with a slow pan.
• The work: a stylist's hands with scissors or a blow-dryer, or a nail tech at the table.
• The results: finished cuts, colors or nails from real clients who agreed to appear.
• The offer: a new-client price, a service of the month, how to book, said in the voice-over.

Can I use AI hairstyle images as before-and-after results?

No. A cut or color changed by AI never happened in your chair, so presenting it as a result misleads the people you want to book. An AI hairstyle changer is fine for showing a client a possible look during a consultation; keep it out of your results. Ask clients before you use their photos in an ad; this post is not legal advice on consent.

Watch the animated clips too: generated motion can change hair length, color or texture. POST /v1/video-inspect returns 8 stills across a finished clip by default, unbilled, so you can compare each one with the photo and drop any shot that drifts.

How do I turn salon photos into video clips?

Put each photo at a public HTTPS URL and send it as the first_frame in frame_images on a POST /v1/videos job, with a prompt for one gentle move and aspect_ratio: "9:16". Each model lists its own ratios and clip lengths at GET /v1/videos/models. Image-to-video API: first and last frame shows the request.

How do I add the voice-over, music and captions?

The assembly works as in Restaurant video advertising with AI; for a salon:

• Write the offer as a voice-over with Sume text-to-speech; it becomes the Timeline 1.0 audio spine, and you set audio.duration_seconds (the output length) to match it.
• Generate a music bed with POST /v1/music-router/generate and add it as the render's soundtrack, with duck_db to lower it under the voice.
• Render with POST /v1/timeline-1.0/render: the default output is 1080 × 1920. In current code each clip's own sound is dropped, so the spine and soundtrack are all you hear.
• Burn captions onto the finished render with POST /v1/video-captions; it can caption from the spoken voice-over. Today it refuses videos over 60 seconds or with no audio stream. See How to burn captions onto a video.

How much does a salon promo video cost?

You pay per step from your workspace USD balance: