Whispers in the Void

Inspiration

I was inspired by the idea of combining AI-generated visuals with music. I wanted to create a music video where an AI-generated female character could actually sing the song, giving life to a virtual performer. This project is a step toward exploring how AI can merge visual art and music in a creative way.

What it does

Through this project, I learned: How to generate photorealistic AI images of human characters. How to animate mouth movements and sync them with a pre-recorded song using AI tools. How to combine audio and visuals into a seamless video. I also deepened my understanding of the challenges of timing and expression in AI animation, and the importance of small details in making the character feel alive.

How we built it

Image Creation: I used AI image generation tools to create a realistic Japanese female character. Animation: Using AI lip-sync tools, I animated the character’s mouth to match the song's lyrics. Video Editing: I combined the singing character with background scenes and added my own music and sound effects.

Challenges we ran into

Achieving realistic expressions and lip-syncing was difficult at first. I solved this by fine-tuning the AI parameters and adjusting timing manually. Balancing the visual style with the music mood required several iterations of image generation and editing.

Accomplishments that we're proud of

Successfully generated a photorealistic AI female character. Created a fully synchronized AI lip-sync animation. Produced a complete music video combining AI visuals and custom music.

What we learned

Learned to fine-tune AI parameters for realistic lip-sync. Improved workflow for combining AI visuals with audio. Gained insights into virtual performance creation.

What's next for Whispers in the Void

Explore interactive AI music videos where viewers can choose the song. Experiment with multiple AI characters performing together. Expand to other musical genres and visual styles.

Built With

• hotgengeneralv4
• seedance