Inspiration
JAPA—a Nigerian slang term for migration—represents the dreams, sacrifices, and difficult choices of millions seeking better opportunities abroad. I wanted to capture this deeply emotional story in a music video for artist Heavins, with Hollywood-level cinematic quality. Traditional filmmaking wasn't accessible, but AI tools opened a new path. The goal was to prove that powerful visual storytelling doesn't require massive budgets—just vision, determination, and the right tools.
What it does
This behind-the-scenes video documents the complete creative journey of making the JAPA music video using AI tools. It shows the process from initial concept to final cut: breaking down the story into 80 individual shots, writing detailed prompts using Claude AI, generating each scene with HailuoAI , Hero Shots with LTX Studio, using Elevenlabs for the Voiceover, Google Cloud to host the work flow, and editing everything together in CapCut. The BTS reveals the challenges, iterations, breakthroughs, and creative decisions that transformed text prompts into a cohesive cinematic experience—demonstrating that AI-powered filmmaking is now accessible to independent artists.
How we built it
Pre-Production: I used Claude AI and ChatGPT to develop a comprehensive 80-shot breakdown, writing detailed prompts for each scene with specific camera angles, lighting, emotions, and visual metaphors.
Production: Each shot was generated using HailuoAI and LTX Studios, feeding in the carefully crafted prompts. Some shots worked immediately; others required 5-10 iterations to achieve the right look and feel.
Post-Production: CapCut served as the editing suite where I assembled all 80 clips, added transitions, synchronized to the music's beat, and applied color grading. ElevenLabs provided the opening voiceover for emotional impact.
Enhancement: I experrimented with Luma AI but had issues with consistency
Documentation: Throughout the process, I captured screen recordings of prompt writing, AI generation, and editing timelines to create this behind-the-scenes documentary.
Challenges we ran into
Character Consistency: The biggest challenge was maintaining the same character's appearance across 80 different AI-generated shots. AI models don't inherently "remember" previous outputs. Solution: I created a master character description sheet and referenced it in every single prompt, reducing inconsistency by approximately 70%.
Timing and Rhythm: Getting AI-generated clips to match the song's beat and emotional peaks required precise planning. Solution: I pre-timed every shot duration in my breakdown and used CapCut's beat-sync features for fine-tuning.
The Hero Shot: The paper plane vortex sequence—symbolizing dreams taking flight—took 5 complete iterations before capturing the right energy and visual impact. Each attempt refined my understanding of how to prompt for dynamic motion.
Technical Limitations: AI tools sometimes generated unexpected results or required multiple attempts to match my vision, teaching me patience and adaptability.
Accomplishments that we're proud of
80 Cohesive Shots: Successfully generated and assembled 80 AI shots into a seamless narrative that tells a compelling migration story.
Character Consistency Achievement: Developed a repeatable system for maintaining visual consistency across AI-generated content—a major technical breakthrough.
Cinematic Quality on Zero Budget: Created Hollywood-level visuals without traditional film equipment, crew, or location costs.
Empowering Independent Artists: Gave Heavins a professional-quality music video that can compete with major label productions.
Mastering the Prompt-to-Screen Pipeline: Became proficient in the emerging craft of AI filmmaking, from concept to final delivery.
Documenting the Journey: Created a comprehensive BTS that can inspire and educate other cre
Built With
- capcut
- chatgpt
- claude
- elevenlabs
- hailuo
- ltx
- luma
Log in or sign up for Devpost to join the conversation.