Prompt With Jarvis
Work with meGet the free pack
All guides

DUBYou commented this keyword

VoiceStudio

An open-source desktop app to clone or design voices, dub videos, dictate and make audiobooks, running on your own hardware.

Before you start:
  • macOS, Windows or Linux
  • About 5 GB free disk for the CPU build
  • A GPU helps but is optional
Setup
15 min
Level
Beginner
Stars
54.3k
License
AGPL-3.0
Jump to the install commandView on GitHub

What it does

VoiceStudio is a desktop app for voice work. You can clone a voice from a clean recording, design a new voice by describing it, dub videos with timed speech, dictate with a floating widget, transcribe audio and create audiobooks.

It runs on your own hardware. It also has a local API and an MCP server (a connector that gives AI agents new tools), so agents like Claude Code can use it.

Install it

  1. On macOS or Linux, run the one-command installer below in your terminal.
  2. On Windows, run this in PowerShell instead: irm https://voicestudio.sh/install | iex
  3. Or download the app from the Releases page (.dmg for macOS, .exe for Windows, .AppImage or .deb for Linux).
  4. Open the app and install the model it asks for.

Copy this

Terminal
curl -fsSL https://voicestudio.sh/install | sh

Or let Claude Code install it

  1. Paste the prompt below into Claude Code. It follows the project's agent guide, checks your hardware, and asks before downloading models.
  2. Agents that support skills can also run npx skills add debpalash/VoiceStudio and choose the voicestudio skill.

Install prompt

Prompt
Install the VoiceStudio Electron app on this device and verify it works, following
https://github.com/debpalash/VoiceStudio/blob/main/docs/install/agent.md

Use it

  1. Open Voice cloning.
  2. Choose a voice or add a clean reference recording.
  3. Type your text and generate.

Good to know

NVIDIA GPUs (Windows, Linux) and Apple Silicon get hardware acceleration. Computers without a dedicated GPU work on the CPU, just slower. Intel Macs only run the app interface and need a remote backend. Windows on ARM is experimental.

Only clone voices with permission. The models have their own licenses, so check them before commercial use.

Links

Want the next one first?

One new guide a week, straight to your inbox.

More like this