Voxify | AI Text-to-Speech Platform Built

Voxify | AI Text-to-Speech Platform

Built a production-ready AI voice platform that lets creators, businesses, and developers generate realistic speech audio at scale — fast, clean, and API-ready.

What the platform does

Multi-voice generation with customizable expression and speaking style, organization-based workspace collaboration, real-time audio synthesis with low latency playback, developer-friendly REST APIs, and usage tracking with credit and analytics systems.

What I built

Led end-to-end development across the full stack — dashboard experience in Next.js, Python FastAPI service on Modal (GPU A10G) for TTS inference, Prisma with PostgreSQL for data modeling, Cloudflare R2 for audio storage and delivery, and a typed API layer using tRPC and OpenAPI typegen. Auth and org management handled via Clerk.

Stack

Next.js + TypeScript — Python FastAPI — Prisma + PostgreSQL — Cloudflare R2 — tRPC — Clerk — Tailwind + shadcn/ui — TanStack Query — Zod

The result

A scalable voice infrastructure product, not just a demo. Teams can generate high-quality voiceovers instantly, manage multiple voice profiles, and plug audio generation directly into their products through a clean API. Built for creators, built for developers, built for scale.