Runs on your laptop, desktop, servers, phone, tablets and more
Website · Download · Get started · Docs · Discord · 简体中文 · 日本語 · 한국어 · Русский
Strategic sponsors
Special sponsor
Demo workspace · Default dark theme · Watch the HD tour
 Voice cloning Turn a clean reference recording into speech in a familiar voice. |
 Video dubbing Translate and dub videos with speech timed to the original. |
 Voice design Describe a voice in words and generate it. |
 Local models Choose your engines and download only the models you need. |
More screenshots
 Workspace overview |
 Model packs |
- Dictate & transcribe — A floating dictation widget and local transcription
- Narrate & batch — Stories, audiobooks and queued jobs
- Connect & automate — Local API, MCP, reusable recipes and watch folders
Start with VoiceStudio (default, powered by k2-fsa/OmniVoice), or pick from dozens of other engines in the Model Manager. See the features & engine catalog.
Choose your system, then expand an installation method.
macOS · Homebrew · DMG · installer
Stable · Apple Silicon
macOS 13.3+ · Apple Silicon: local engines with Metal (MPS).
brew tap debpalash/tap brew trust --cask debpalash/tap/voicestudio brew install --cask debpalash/tap/voicestudioOne-command installer
curl -fsSL https://voicestudio.sh/install | shIntel Mac · remote backend · Limited
The app UI runs on Intel; connect it to a VoiceStudio backend on another computer.
[!NOTE] On macOS, "VoiceStudio is damaged and can't be opened" means Gatekeeper blocked the app, not that the download is broken. See the Gatekeeper fix.
Windows · PowerShell · EXE
Stable · x64
Windows 10/11 x64 · NVIDIA CUDA · AMD ROCm (opt-in) · CPU.irm https://voicestudio.sh/install.ps1 | iex
Linux · AUR · Snap · DEB · AppImage · Nix
Stable · x64
Linux x64 · NVIDIA CUDA · CPU · Vulkan for supported GGUF engines.One-command installer
curl -fsSL https://voicestudio.sh/install | shArch / Manjaro / EndeavourOS · AUR
yay -S voicestudio-binAlso available through
paru -S voicestudio-bin.Development sources ·
yay -S voicestudio-gitBuild from AUR and install with pacman
sudo pacman -S --needed base-devel git git clone https://aur.archlinux.org/voicestudio-bin.git cd voicestudio-bin makepkg -s sudo pacman -U ./voicestudio-bin-*.pkg.tar.zstUbuntu / Debian / Linux Mint · DEB
Download the
.debinto the current folder, then install it:sudo apt install ./VoiceStudio-*-linux-x64.debSnap · Ubuntu, Fedora, openSUSE and more
With snapd installed:
sudo snap install voicestudioTo enable microphone access:
sudo snap connect voicestudio:audio-recordAny Linux distribution · AppImage
Download the AppImage into the current folder, then launch it:
chmod +x VoiceStudio-*-linux-x64.AppImage ./VoiceStudio-*-linux-x64.AppImageNix / NixOS · source environment
Enter the repository’s FHS development environment, then run from source:
git clone https://github.com/debpalash/VoiceStudio.git cd VoiceStudio nix develop .#fhs bun run dev
Docker · GHCR · Docker Hub
Stable images · CPU · NVIDIA CUDA · AMD ROCm.
Run locally on the CPU (Bash)
export OMNIVOICE_API_KEY="$(openssl rand -hex 32)" docker run -d --name omnivoice \ -p 127.0.0.1:3900:3900 \ -e OMNIVOICE_API_KEY \ -e OMNIVOICE_DATA_DIR=/app/omnivoice_data \ -v omnivoice-data:/app/omnivoice_data \ ghcr.io/debpalash/voicestudio:stable printf 'API key: %s\n' "$OMNIVOICE_API_KEY"Open http://localhost:3900 and use the API key printed below.
NVIDIA, AMD ROCm and Compose instructions
Docker Hub mirror:
palashdeb/omnivoice-studio:stable
Android · APK · Obtainium · F-Droid repository
Android 10+ · Preview · On-device, offline after model download.
F-Droid · VoiceStudio repository
Add this repository in an F-Droid-compatible client; verify the fingerprint in the guide.
https://debpalash.github.io/VoiceStudio/fdroid/repoObtainium · GitHub releases
Add this source in Obtainium, or import the provided configuration.
https://github.com/debpalash/VoiceStudioInstall over USB with adb
adb install -r VoiceStudio-*-android.apk
iPhone / iPad · IPA · sideloading
iOS / iPadOS 17+ · Preview · On-device, offline after model download.
Download the IPA and follow the signing and sideloading instructions.
Desktop and phone installers are on the latest release; Docker images come from GHCR. No GPU? The app is fully usable on the CPU, just slower. Native GGUF engines can also use Vulkan on AMD, Intel and NVIDIA GPUs; OmniVoice GGUF picks its GPU and model size automatically on Windows and Linux (OmniVoice, CosyVoice). Speed by engine and hardware: performance.
Install using the options above. Then open Voice cloning, pick a voice or add a clean reference recording, type your text and generate. The app asks before it downloads any model.
More install options: a specific version, current main, uninstall, from source, with a coding agent
curl -fsSL https://voicestudio.sh/install | sh -s -- --version X.Y.Z # a specific release
curl -fsSL https://voicestudio.sh/install | sh -s -- --main # build current main
curl -fsSL https://voicestudio.sh/install | sh -s -- --uninstall # remove the app, keep your dataIf voicestudio.sh is unreachable, run the script straight from the repository:
curl -fsSL https://raw.githubusercontent.com/debpalash/VoiceStudio/main/scripts/install.sh | sh.
Prerequisites and behavior: installer guide.
From source
git clone https://github.com/debpalash/VoiceStudio.git
cd VoiceStudio
bun run devbun run dev installs the JavaScript dependencies and, when using a supported
local backend, the Python dependencies before launching Electron. Run
bun run setup:api directly when you only need to rebuild the Python environment.
See Electron setup.
With a coding agent (Claude Code, Codex, Cursor, …), paste:
Install the VoiceStudio app on this device and verify it works, following
https://github.com/debpalash/VoiceStudio/blob/main/docs/install/agent.md
Upgrading from 0.5.3 or older (the Tauri app)? Install the new app separately.
| Need | Start here |
|---|---|
| Setup help | Troubleshooting · Model downloads |
| Guides | Multi-speaker dubbing · Expressive speech · Exporting dubs |
| Models & quality | Engine guides · whisper.cpp ASR · Benchmarks |
| Integrations | Local API · MCP · Examples |
| Phone apps | iOS and Android |
| Development | Contributing (sign the CLA before your first PR) · Changelog |
Agent skills: npx skills add debpalash/VoiceStudio (voicestudio for audio work, voicestudio-maintainer for the repo).
VoiceStudio is free and has no paid cloud behind it. Sponsors keep it shipping: they fund maintainer time, test hardware, code-signing and release infrastructure.
For sponsorship inquiries, contact sponsor@voicestudio.sh.
More open-source tools from the people building VoiceStudio.
- OpenGTM — Find and enrich companies and people in a self-hosted go-to-market workspace. Source.
- Opal — Watch, stream and browse your media in one native, open-source player. Source.
- Supermetry — Understand your websites and products with self-hosted, cookie-free analytics. Source.
More projects: debpalash · velixio
 Palash Debnath Creator and Core Team palash.dev · GitHub · X |
 Shimul Debnath Core Team shimul.me · GitHub · X |
Thank you to everyone who sends code, reports, translations and ideas. All contributors · Contribute
AGPL-3.0 covers the app, including commercial use under its terms. Model weights, voices and generated output have their own licences, which the Model Manager shows for each model (inventory). Clone a voice only with its owner's permission. Details: LICENSE-NOTICE.md.