voicestudio.sh

voicestudio.sh is an open-source desktop application that puts a full voice AI studio on your own computer. It can clone a voice from a short recording, design a brand-new synthetic voice from a text description, dub videos into different languages, transcribe audio, and generate audiobooks — all without sending any data to the cloud. The tool supports 646 languages and runs entirely offline, making it a strong local alternative to services like ElevenLabs.

The app is built around a set of dedicated workspaces. The Voice Clone screen lets you record or upload a sample and create a voice in minutes. Voice Design works the other way: you describe the voice you want (gender, age, accent, pitch) and the tool generates it from scratch. The Dubbing editor handles the full pipeline — transcribe the source audio, translate the text, assign speakers, generate new audio, and export. There's also a Voice Gallery with ready-made voices, a Studio workspace for fine-tuning scripts and styles, and a local API that's compatible with OpenAI's format so developers can integrate voice AI into their own projects. Everything runs on your machine using the OmniVoice local engine.

voicestudio.sh is ideal for content creators who need voiceovers and dubbing without monthly fees or privacy concerns. Developers can use the local API to build voice features into their apps. Teams can purchase commercial licenses or hosted access for larger projects. The tool is free for personal use, and because it's open source, you can inspect exactly how it works.