VoiceStudio

VoiceStudio is an open-source, fully local alternative to ElevenLabs for voice cloning, voice design, video dubbing, dictation, transcription and audiobook creation, with more than 79,000 stars on GitHub. Everything runs on your own machine, so your audio never has to leave your computer.

Overview

Cloud voice platforms charge per character and require you to upload recordings to someone else’s servers. VoiceStudio takes the opposite approach: it is a desktop application (Electron, for macOS, Windows and Linux, plus Docker) that bundles local speech models and runs them on your hardware, with CUDA acceleration on NVIDIA GPUs, Metal on Apple Silicon, and a fully usable CPU fallback. It ships with the OmniVoice engine by default and lets you install and switch between multiple speech engines from its model manager.

The workspace is organized around complete production workflows rather than a single text box: clone a voice from a clean reference recording, design a new voice from a description, dub videos with timed speech, dictate through a floating widget, or turn long text into stories and audiobooks with batch jobs. Support for 646 languages makes it unusually broad for a local tool.

Key Features

• Voice cloning and design – clone a voice from a reference recording or describe the voice you want and generate it.
• Video dubbing – dub videos with timed speech matched to the original footage.
• 646 languages – speech generation and transcription across an exceptionally wide language catalog.
• Fully local processing – no account and no cloud; it only goes online to check updates or download models you choose.
• Dictation and transcription – a floating dictation widget plus speech-to-text for recordings and batch jobs.
• Audiobook and story production – long-form narration workflows with reusable recipes and watch folders.
• Local API and MCP – connect agents and other tools to VoiceStudio through its local API and MCP support.

💳 Pricing

VoiceStudio is free and open source under the AGPL-3.0 license, including self-hosted and desktop use. An optional closed-source Pro build is offered for teams that need commercial license terms, priced at $99/year or a $299 lifetime option; the open-source app itself costs nothing.

Who Is It For

Creators, podcasters, video producers, narrators and developers who want ElevenLabs-style voice workflows without per-character cloud bills or uploading private audio, and who are comfortable running models on their own hardware.