How to turn photos into a Reel with a voice-over and captions

Upload the photos to SPREVA, say what the video is about, and choose the format, the length and a voice. SPREVA writes the script, records the voice, puts each photo and caption on the words it belongs to, and opens the result in Studio. Change what you want by asking, export it, and schedule it to Instagram, TikTok and YouTube, each in its own format.

• Five to ten photos or short clips: JPEG, PNG, WebP, AVIF or GIF, or video up to 1 GB. HEIC is not accepted yet, so share iPhone photos as JPEG.
• A plan with Studio: Pro or Business. The 7-day trial includes 300 AI credits and 10 minutes of exports.
• The accounts you want to post to, connected to SPREVA. Instagram publishes only to professional accounts.

1. 1

   Upload the photos to Media. Each one is described as it arrives, so you can find it later by what it shows.

2. 2

   In Studio, choose New project, then From a script or an idea, and start From an idea.

3. 3

   Say what the video is about and who it is for, add any facts it should use under Notes, and pick 9:16 and a length.

4. 4

   Under Pictures, choose your photos, or a folder of them: each line of the script gets the one that shows what it says. Then pick music from your library, or none, and a voice.

5. 5

   Choose Make the first cut. SPREVA writes the script, records the voice and builds the video with captions timed to every word, and it opens in Studio. It says how many AI credits it may use before it starts.

For a video set to music, choose Create with AI in Studio instead: write a line about the video, add your photos and clips, and set the format, the length and the pace, calm, balanced or fast. It builds an editable first cut on the timeline.

With your agent connected to SPREVA, ask for the whole thing in the chat:

The agent starts a project, writes the script, asks you before the voice-over uses AI credits, places each photo and caption on its words, checks the cut and exports it. Then ask for changes the way you would to a person:

Only what you named changes: the voice, the captions and the other scenes stay where they were. It works the same in Claude, ChatGPT, Gemini or any agent that speaks MCP.

The video is a Studio project, so nothing has to be made again:

• Open the AI beside the editor and ask in plain words: “warmer colours on every shot”, “add a dissolve between shots”. Each request is one step you can undo.
• Or change it on the timeline: move a photo, trim a scene, restyle the captions, add your logo.
• Change the script later and the video can follow, scene by scene, and the narration can be recorded again.
• Export it as an MP4 into your library, ready for a post.

Put the exported video in a post, pick Instagram, TikTok and YouTube, and give any of them its own caption. It goes out as a Reel, a TikTok video and a Short. Publish now, on a date or through a queue. How to adapt one post for each network.