Overview

Duramo Lab is a production AI voice platform that converts text to expressive speech and clones voices from short samples. Paste a script, choose a voice, and generate an audio result in seconds. The platform runs on a versioned REST API with JWT authentication, making it easy to embed into any application stack.

The product launched with support for multiple languages, a gallery of production voices, and a self-serve credit system: 35,000 free credits on signup, no card required. It serves content creators, e-learning platforms, developers, marketing teams, and enterprise clients across five continents.

The Challenge

Building a TTS platform that actually sounds human is a solved problem at the infrastructure level, but the UX around it is hard to get right. The challenge was building an end-to-end product: speech synthesis engine, voice cloning pipeline, developer API, credit system, and a polished interface, all production-grade from day one. The API had to be clean enough for a solo developer to integrate in an afternoon, while the infrastructure had to handle enterprise load without downtime.

The Approach

I built Duramo Lab as a full-stack product with a Next.js frontend, Python backend handling speech synthesis, and PostgreSQL for persistence. The API is versioned and documented: JWT auth, usage tracking, rate limits, and authenticated media delivery are all built in.

Voice cloning works from short consented samples. Users upload audio, the pipeline extracts voice characteristics, and the cloned voice becomes available for text-to-speech generation. Speed, pitch, and emotion controls give fine-tuned output without overwhelming the user.

The credit system was designed to be generous at the free tier (35,000 credits, no card) while scaling cleanly through paid plans. Credits map to character counts, making usage transparent and predictable for teams budgeting voiceover costs.

Highlights

Text-to-speech with multiple production languages and expressive output
Consented voice cloning from short audio samples
Versioned REST API with JWT authentication and usage tracking
35,000 free credits on signup, no credit card required
Pricing from free through enterprise, serving solo creators to large teams
Tested and endorsed by Markaba Studios, Xelaware, HNI, Cedar Labs, and others

The Outcome

Duramo Lab is live and growing, serving content creators, e-learning platforms, and enterprise teams across five continents. The API runs inside IVR systems, podcast pipelines, documentary production, and accessibility tools. The product proves a small team can ship AI infrastructure that competes with players ten times its size.