Today, I’m showing you how to take a single static image and turn it into a short storytelling video with sound effects and character acting. We’ll use Midjourney for the look, Higgsfield for the motion, and ElevenLabs for the voice.
Step 1: Create the Character (Midjourney)
Go to Midjourney and generate your base image.
The Prompt:
"A black girl with dreadlocks leaning on a rock next to her cub lion, they appear to be stargazing. It's sunset. There's a city skyline of a future punk city."
Download your favorite result.
Step 2: Animate the Scene (Higgsfield)
Go to Higgsfield.ai, click on Video in the menu, and select Seedance (this is the model we’ll use).
Upload your Midjourney image.
The Prompt:
"The lion roars then the girl roars. They then look at each other and smile. The frame is locked."
Note: Adding "The frame is locked" is crucial to stop the camera from cutting away or panning weirdly.
Download the generated video.
Step 3: Generate Lion Sound Effects (Adobe Firefly)
Go to Adobe Firefly and look for the Audio or Sound Effects generation tool.
The Prompt: "Cub lion roaring a long roar."
Generate a few options and download the best one (look for one that matches the duration of your video).
Step 4: Create the "Girl Roar" (ElevenLabs)
Go to ElevenLabs and select Voice Changer.
Record yourself: Record yourself making a roaring sound into your microphone (try to match the timing of the girl’s mouth in the video).
Choose a Voice: Select a character voice (I used a custom voice, but you can search for "Lollipop" or any "Girl" voice in the library).
Hit Generate to transform your roar into her voice.
Step 5: Final Mix (CapCut)
Bring your video, the lion sound, and the converted voice recording into CapCut.
Sync the audio to the visuals.
And that’s it! You’ve just created a fully animated scene with custom acting and sound design from scratch.
