# Duramo Lab - AI Voice Platform (Full Technical Documentation)

## Overview
Duramo Lab is an AI text-to-speech and voice cloning platform. Built for creators, developers, and enterprises. It uses Inworld AI's TTS engine to deliver high-quality, expressive speech synthesis with real-time voice cloning.

## Core Features

### Text-to-Speech (TTS)
- Turn text into natural speech
- Supported production languages are returned by `/api/v1/languages/`
- Adjustable speed, pitch, and emotion
- Chunking for long-form text (5000+ characters)
- Async processing with queue management
- Real-time progress tracking

### Voice Cloning
- Clone an authorized audio sample within the documented upload duration limits
- Async processing with account and plan limits
- WAV, MP3, M4A, OGG formats supported
- Quality validation and rejection
- Custom voice names and descriptions

### Voice Gallery
- Pre-built voices ready to use
- Filter by language, gender, emotion
- Preview before you use

### API Access
- RESTful API for programmatic access
- JWT authentication
- Rate limiting and quota management

## API Endpoints

### Authentication
- POST /api/v1/auth/register/ - Create account
- POST /api/v1/auth/login/ - Get JWT tokens
- POST /api/v1/auth/refresh/ - Refresh access token
- POST /api/v1/auth/logout/ - Blacklist tokens
- GET /api/v1/auth/profile/ - Get user profile
- PATCH /api/v1/auth/profile/update/ - Update profile
- POST /api/v1/auth/change-password/ - Change password

### TTS
- POST /api/v1/tts/synthesize/ - Generate speech (async)
- GET /api/v1/tts/history/ - List generations
- GET /api/v1/tts/history/<id>/ - Get generation details
- GET /api/v1/tts/history/<id>/status/ - Get task status
- GET /api/v1/tts/history/<id>/download/ - Download audio

### Voices
- GET /api/v1/voices/ - List user voices
- POST /api/v1/voices/upload/ - Upload voice for cloning
- DELETE /api/v1/voices/<id>/ - Delete voice

### Available Voices
- GET /api/v1/tts/voices/ - List available voices

### Projects
- GET /api/v1/projects/ - List projects
- POST /api/v1/projects/ - Create project
- GET /api/v1/projects/<id>/ - Get project details
- PUT/PATCH /api/v1/projects/<id>/ - Update project
- DELETE /api/v1/projects/<id>/ - Delete project

## Pricing Tiers

### Free
- Free credits: 35,000
- Voice clones: 0 in the current seeded Free plan
- TTS retention: 24 hours
- Max chars per request: 5,000

### Pro
- Higher character limits
- More voice clones
- Extended TTS retention
- Priority queue processing

### Enterprise
- Custom limits
- Dedicated support
- No contractual SLA is promised by the self-serve plan

## Tech Stack

### Backend
- Django 5.x
- Django REST Framework
- Django REST Framework SimpleJWT

### Async Processing
- Celery
- Redis (broker + cache)
- Separate worker pools for TTS and voice cloning

### Database
- PostgreSQL (Neon)
- Connection pooling with conn_max_age

### TTS Engine
- Inworld AI TTS API
- 3-tier voice fallback (cloned > special > builtin)

### Infrastructure
- Hostinger VPS (2 CPU, 8GB RAM, Ubuntu 24.04)
- Gunicorn (gthread, 16 workers)
- Nginx reverse proxy
- Sentry error tracking

## Quick Start Guide

1. Sign up at https://duramolab.com/signup/
2. Verify your email
3. Go to Dashboard > TTS Studio
4. Select a voice or upload your own
5. Enter text and click Generate
6. For API access, obtain a JWT through the documented login endpoint
7. Use the JWT only in the Authorization header

## Contact
- Website: https://duramolab.com
- Email: support@duramolab.com