VoiceStudio

VoiceStudio is a desktop app that runs AI voice tools on your own device: voice cloning from a reference recording, voice design from a text description, video dubbing with timed speech, dictation, transcription and audiobook narration, in hundreds of languages. It is aimed at creators who want an ElevenLabs-style workflow without uploading audio to a cloud service.

You pick and download only the speech engines you need from a built-in model manager, and it can use CUDA, Apple MLX or Vulkan GPUs. It also offers a floating dictation widget, batch jobs, a local API and MCP. The open-source app is free; an optional paid Pro edition adds a commercial license and multi-device features.