Khari Clemens 👋
Founder of Helexeon, LLC.
Helping artists, entrepreneurs, and innovators harness the power of generative AI for storytelling, branding, and automation.
Project:
Fuel Injected Passion
Music Video
Overview
Another chapter in the upcoming short film, "The Prototype". Avva (one of our 2 main characters) breaks free from the digital mainframe, blazing through neon streets, transforming from machine to motorcycle to sports car to starship in a relentless chase for freedom. Set to an adrenaline-charged track, it’s a cyberpunk spectacle of speed, power, and transcendence — hinting at a galaxy that may be one vast simulation. This is another chapter from an upcoming short film, expanding the story of Devvin and Avva’s intertwined fates.
Concept: To create a music video as part of a sequence to further the story of the upcoming short film, "The Prototype".
Planning: The story is set in a cyberpunk world, so we needed to create a visual setting that would be consistent with the story. Storyboard and concept art were mainly brainstormed in chatGPT then carried out through other software platforms.
Tools Used: It took about a dozen tools to create the final project. Included:
- ChatGPT (Brainstorming and Concept Development)
- Midjourney (Image and Video generation)
- Kling AI (Image and Video generation)
- VEO 3 (Video generation)
- Adobe Photoshop (Image editing)
- Adobe Premiere Pro (Video editing)
- HeyGen (Video creation, mainly for lip sync)
- Apple Logic Pro (Music Production)
- Suno (Vocal Stems)
- HiggsField (Video editing, mainly for transitions)
- Microsoft Word (Narrative and Dialogue Writing)
- ElevenLabs (Voiceovers and Narrations)
Challenges
You may often see a lot of AI generated images and videos, but it's not always easy to tell the difference. One thing that makes this video standout is the use of consistent characters in multiple scenes. Also the use of lip sync, which many music videos fail to include. Here are some of the challenges overcome:
Character Consistency:
- Challenge: Creating a consistent character design that can be used in multiple scenes.
- Solution: Use of OmniReference and voice training to create a consistent character design and sound. Training the AI to use the same character design in multiple scenes.
Lip Sync / Voice Training:
- Challenge: Lip sync and dancing on beat.
- Solution: Use of HeyGen to create lip sync and post-editing to ensure the dancing is on beat. AI doesn't know what the song or BPM is, so editing is required to ensure the dancing is on beat. Also, the AI doesn't know the rhythm or pace of the music, so careful editing was required to achieve the desired effect.
Video Editing / Transitions:
- Challenge: Creating a seamless and intuitive video editing and transitions.
- Solution:Many AI generated videos consist of random 5 or 10 second clips without continuity across them. We used start and end frames mixed with clever transitions to achieve a professional result
Compelling Storytelling:
- Challenge: Creating a compelling story that is engaging and keeps the viewer interested.
- Solution: We used a combination of AI generated images and videos, and carefully crafted the narrative to create a compelling story.
Results/Conclusion:
The video was well received by the target audience and the attention to detail was commended by viewers and fellow artists. The goal of creating something more than just a random mix of visuals was accomplished. View it now on YouTube, or above in the video embed.
