SUZA Voice Studio 2.0.2 Docs - Install on Windows, macOS, and Linux
Version 2.0.2 installation docs

Install SUZA Voice
Studio on your desktop.

Follow the platform guide for Windows, macOS, or Linux. The packaged installers do not require Python; you only need an internet connection for first-run model downloads.

Start install guide Back to homepage
Quick Start

Pick the installer for your operating system.

Gumroad handles payment and file delivery. After purchase, download the installer for your platform, install the app, then open SUZA Voice Studio 2 and approve the first model download when prompted.

Windows

Download the SUZA_Voice_Studio_2.0.2_Setup.exe installer and run it.

macOS

Download the SUZA_Voice_Studio_2.0.2.dmg, open it, and drag the app to Applications.

Linux

Use the AppImage for portable launch, or install the Debian package on Debian/Ubuntu systems.

Requirements

What you need before installing.

  • Windows 10/11 64-bit, macOS 13 or newer, or a modern 64-bit Linux desktop.
  • Internet access for the first model download and optional sample/catalog refreshes.
  • Several GB of free disk space for voice models. Qwen models are larger than Piper voice models.
  • For Fast Mode, a normal CPU is enough. For Quality Mode, an NVIDIA CUDA-capable GPU is recommended; otherwise Qwen can fall back to CPU and will run slower.
Windows

Install on Windows.

  1. Download the Windows installer from your Gumroad library.
  2. Run SUZA_Voice_Studio_2.0.2_Setup.exe.
  3. Follow the installer prompts and keep the Desktop/Start Menu shortcut enabled if you want quick access.
  4. Open SUZA Voice Studio 2 from the Start Menu.

Optional: check NVIDIA GPU availability

If you plan to use Qwen3-TTS Quality Mode with GPU acceleration, run this in PowerShell:

PowerShell
nvidia-smi

If the command is not found or cannot see your GPU, install or update the NVIDIA driver, then restart the PC.

macOS

Install on Mac.

  1. Download the Mac DMG from your Gumroad library.
  2. Open SUZA_Voice_Studio_2.0.2.dmg.
  3. Drag SUZA Voice Studio 2.app into the Applications folder.
  4. Open the app from Applications.

If macOS blocks the app

If Gatekeeper says the app cannot be opened, right-click the app in Applications and choose Open. If you still need to remove the quarantine flag, run:

Terminal
xattr -dr com.apple.quarantine "/Applications/SUZA Voice Studio 2.app"
open "/Applications/SUZA Voice Studio 2.app"
GPU note for Mac

Fast Mode works locally on Mac. Qwen Quality Mode can fall back to CPU if GPU acceleration is unavailable on the machine.

Linux

Install on Linux.

Linux downloads include an AppImage and a Debian package. Use the Debian package on Ubuntu/Debian if you want system integration; use the AppImage if you prefer a portable executable.

Option A: AppImage

Terminal
chmod +x SUZA_Voice_Studio_2.0.2_x86_64.AppImage
./SUZA_Voice_Studio_2.0.2_x86_64.AppImage

Option B: Debian package

Ubuntu / Debian
sudo apt update
sudo apt install ./suza-voice-studio_2.0.2_amd64.deb

AppImage dependencies on Ubuntu/Debian

If the AppImage does not open, install the desktop runtime libraries:

Ubuntu / Debian
sudo apt update
sudo apt install -y libgl1 libegl1 libnss3 libdbus-1-3 libfontconfig1 libx11-xcb1 libxrender1 libxcb1 libxcb-cursor0 libxcb-glx0 libxcb-icccm4 libxcb-image0 libxcb-keysyms1 libxcb-randr0 libxcb-render0 libxcb-render-util0 libxcb-shape0 libxcb-shm0 libxcb-sync1 libxcb-xfixes0 libxcb-xinerama0 libxcb-xkb1 libxkbcommon-x11-0 libpulse-mainloop-glib0 sox libfuse2
sudo apt install -y libasound2t64 || sudo apt install -y libasound2

Fedora

Terminal
sudo dnf install -y mesa-libGL mesa-libEGL nss dbus-libs fontconfig libX11 libXrender libxcb xcb-util-cursor xcb-util-image xcb-util-keysyms xcb-util-renderutil libxkbcommon-x11 alsa-lib pulseaudio-libs sox fuse-libs

Arch Linux

Terminal
sudo pacman -S --needed libglvnd nss dbus fontconfig libx11 libxrender libxcb xcb-util-cursor xcb-util-image xcb-util-keysyms xcb-util-renderutil libxkbcommon-x11 alsa-lib libpulse sox fuse2

Optional: check NVIDIA GPU availability

Terminal
nvidia-smi
First Run

Download models only when you choose them.

SUZA Voice Studio does not bundle voice model files in the installer. On first synthesis or first voice selection, the app asks for permission to download the selected model, then caches it locally for future offline use.

  1. Open SUZA Voice Studio 2.
  2. Choose Fast Mode for Piper TTS or Quality Mode for Qwen3-TTS.
  3. Select a language, voice, or Qwen profile.
  4. Approve the model download prompt.
  5. Generate audio. Future runs can use the cached model without downloading again.
Data Locations

Where SUZA stores models and settings.

Generated audio, downloaded models, samples, clone profiles, and settings are stored locally on your machine.

  • Windows: %LOCALAPPDATA%\SUZA Voice Studio 2\
  • macOS: ~/Library/Application Support/SUZA Voice Studio 2/
  • Linux: ~/.local/share/SUZA Voice Studio 2/
Troubleshooting

Common installation fixes.

The app opens, but model download fails.

Check your internet connection, VPN/firewall rules, and available disk space. Then reopen the app and select the model again.

Qwen Quality Mode is slow.

Quality Mode is fastest with a CUDA-capable NVIDIA GPU. If CUDA is unavailable, SUZA can fall back to CPU inference, which is much slower.

The Linux AppImage does not start.

Install the runtime libraries listed in the Linux section, make sure the AppImage is executable, and try launching it from Terminal to see the exact error.

macOS says the app is from an unidentified developer.

Right-click the app in Applications and choose Open. If needed, run the quarantine removal command in the macOS section.