
Overview
Full-stack platform with audio pipelines, FFmpeg video generation, Runway AI video, AI image fallbacks, and Google speech-to-text/text-to-speech.
Problem
Artisans need a way to share stories and products at scale without producing their own video/audio content from scratch.
What I built
Next.js + Prisma + Postgres on AWS; audio handled through S3 + Lambda pipelines; FFmpeg automates video generation, with Runway AI for enriched video and AI image fallbacks when assets are missing; Google speech-to-text and text-to-speech bridge audio and interface.
Architecture
Hover a node for a short note on what I built there. Dashed edges are async or secondary paths.
Screenshot
