Inspiration
We wanted to explore absurdity and grotesque humor through AI, focusing on a character exaggerating his beauty and intellect. The project demonstrates how multiple AI tools can collaborate to create a fully synthetic, expressive narrative.
What it does
The episode presents a self-absorbed monologue in a mock-interview style, blending humor and unease. AI-generated visuals, voice, music, and text come together to form a cohesive, immersive experience that highlights both the comedic and unsettling qualities of the character.
How we built it
Images: Generated with Midjourney to create expressive character visuals and environments.
Video: Animated and composed with Hailuo.
Lipsync: Managed using Hedra for precise mouth movement.
Voice: Delivered by ElevenLabs, producing natural, expressive narration.
Music: Composed in Udio to match mood and pacing.
Text/Dialogue: Written with ChatGPT to provide witty, exaggerated monologue lines. All elements were carefully synchronized for timing, rhythm, and emotional impact.
Challenges we ran into
Ensuring lip-sync accuracy, maintaining visual coherence across AI-generated frames, and balancing humor with a slightly eerie tone required multiple iterations.
Accomplishments
We created a fully AI-generated, character-driven narrative that feels expressive, coherent, and immersive. The combination of image, sound, and text delivers both humor and tension simultaneously.
What’s next
Expand the series with new AI-generated characters, longer sequences, and explore interactive or immersive AI storytelling possibilities.
Built With
- elevenlabs
- hailuo
- hedra
- midjourney
- udio
Log in or sign up for Devpost to join the conversation.