The story of
Baby Designers
Brand & Creative
Gen AI
Strategy & Growth
Brand & Creative
Ideation
Concept Development
UX & Design
Gen AI
AI Trailers
Prompt Engineering
AI Rapid Concepting
Exploring an end-to-end AI-driven workflow for creating consistent, character based podcast and video content.
Exploring an end-to-end AI-driven workflow for creating consistent, character based podcast and video content.
I created a short “baby podcast” featuring two designer-inspired AI characters, using a streamlined, multi-tool workflow. The characters and script were generated in ChatGPT with the new 4o-image model, enabling a strong visual and narrative foundation. I then produced the character voices with ElevenLabs, an advanced text-to-speech platform known for highly realistic, emotionally expressive voice synthesis across multiple languages and styles.
Next, I processed the audio and character image references through the Hedra Character 3 model, which enabled the creation of expressive, lip-synced character animations from static inputs. Background music was generated with Riffusion, an open-source AI tool that converts text prompts into original audio using spectral image synthesis. Final editing and integration were completed in DaVinci Resolve. This workflow offered an efficient and cohesive method for producing AI-driven, character-based audio-visual content.
I created a short “baby podcast” featuring two designer-inspired AI characters, using a streamlined, multi-tool workflow. The characters and script were generated in ChatGPT with the new 4o-image model, enabling a strong visual and narrative foundation. I then produced the character voices with ElevenLabs, an advanced text-to-speech platform known for highly realistic, emotionally expressive voice synthesis across multiple languages and styles.
Next, I processed the audio and character image references through the Hedra Character 3 model, which enabled the creation of expressive, lip-synced character animations from static inputs. Background music was generated with Riffusion, an open-source AI tool that converts text prompts into original audio using spectral image synthesis. Final editing and integration were completed in DaVinci Resolve. This workflow offered an efficient and cohesive method for producing AI-driven, character-based audio-visual content.
