Skip to main content
Resonance is an open-source text-to-speech and voice cloning platform built with modern web technologies. It provides professional-grade voice synthesis powered by Chatterbox TTS, with a clean and responsive interface built on Next.js 16 and React 19.

​
What is Resonance?

Resonance is a self-hostable alternative to commercial TTS services like ElevenLabs. It combines cutting-edge AI voice synthesis with a production-ready web application that includes authentication, billing, and multi-tenant organization support. The platform leverages serverless GPU infrastructure to provide cost-effective, pay-per-use text-to-speech generation without the overhead of managing ML infrastructure.

​
Key Features

Text-to-Speech

Generate natural-sounding speech from text with adjustable creativity, variety, expression, and flow parameters. Fine-tune temperature, top-p, top-k, and repetition penalty for precise control.

Zero-Shot Voice Cloning

Upload or record a voice sample (10s minimum) and clone it instantly. No fine-tuning or training required - the model adapts immediately to new voices.

20 Built-in Voices

Pre-seeded system voices across 12 categories including Audiobook, Conversational, Narrative, Characters, Meditation, Podcast, and more. Supports 5 locales (en-US, en-GB, es-ES, fr-FR, de-DE).

Multi-Tenant Architecture

Team-based access via Clerk Organizations with full data isolation. Each organization has separate voice libraries and generation history.

Usage-Based Billing

Pay-as-you-go character metering integrated with Polar. Track usage in real-time and configure custom pricing tiers.

Generation History

Browse and replay all past generations with preserved voice metadata and synthesis parameters. Never lose your audio outputs.

​
Technology Stack

Resonance is built with best-in-class open-source technologies:
  • Next.js 16 - React framework with App Router
  • React 19 - Latest React with concurrent features
  • tRPC - End-to-end typesafe APIs
  • Prisma - Type-safe database ORM with PostgreSQL
  • Chatterbox TTS - Open-source zero-shot voice cloning model by Resemble AI
  • Modal - Serverless GPU infrastructure
  • Cloudflare R2 - S3-compatible object storage
  • Clerk - Authentication and organization management
  • Polar - Usage-based billing and metering

​
Use Cases

Generate voiceovers for videos, podcasts, and audiobooks. Clone your own voice or use pre-built voices to maintain consistency across content.
Convert written content to audio for visually impaired users or create audio versions of documentation and articles.
Quickly test voice interfaces and conversational AI applications without recording audio.
Create voice content in multiple languages and accents using the built-in locale support.
Generate audio learning materials, language practice content, or narration for educational videos.

​
Open Source

Resonance is fully open-source and available on GitHub. It’s designed to be self-hosted on your own infrastructure, giving you complete control over your data and costs.
Resonance is also available as a free 12-hour video course that teaches you how to build the entire project from scratch. Each chapter has a corresponding Git branch for easy reference.

​
Getting Started

Ready to get started? Check out the Quickstart guide to run Resonance locally, or dive into the Architecture documentation to understand how everything works together.

Build docs developers (and LLMs) love

Get started for freeTalk to us