Skip to main content

Uberduck API Overview

Welcome to the Uberduck API documentation. Our API provides a unified interface to access a wide range of AI models for speech synthesis, voice cloning, and music generation. This document outlines the core concepts, endpoints, and usage patterns.

Base URL​

All API requests should be made to the following base URL:

https://api.uberduck.ai/v1

Core Concepts​

Capabilities​

The API is organized around capabilities - high-level functions that models can perform:

  • text-to-speech - Convert text to spoken audio
  • voice-cloning - Create voice models from audio samples
  • music-generation - Create music from text descriptions

Models and Providers​

Models are implementations of capabilities, typically from a specific provider:

  • Provider - The source of the model (e.g., AWS, Google, Azure)
  • Model - The specific implementation (e.g., Polly, WaveNet, Neural)

Voices​

Voices are used with speech models and can be:

  • Provider Voices - Pre-built voices from providers (e.g., AWS Polly's "Joanna")
  • Zero-Shot Voices - Custom voices created from user audio samples
  • Fine-Tuned Voices - Extensively trained custom voice models

Available Endpoints​

EndpointDescription
GET /v1/modelsList available TTS models
GET /v1/voicesList available voices with filtering options
POST /v1/text-to-speechGenerate speech from text using specified voice and model

Getting Started​

To start using the Uberduck API:

  1. Obtain your API key
  2. Make your first request
  3. Explore available voices and models

API Versioning​

The current version is v1. We follow semantic versioning principles for all API changes:

  • Major versions (v1, v2, etc.) may include breaking changes
  • Minor updates are backward-compatible and introduce new functionality
  • Patch updates are backward-compatible bug fixes

Major version changes will be announced with a minimum of 6 months' notice before the previous version is deprecated.