Speech host
No strict minimum is documented; a modern multi-core CPU helps, and more RAM helps if you pick larger STT/LLM models. CPU-optimized int8 STT works without a dedicated GPU.
Test before you buy
Know what is supported on your Linux system.
Install OSS and verify the exact microphone, display session, target applications, text-insertion backend, and model endpoint you intend to use. Pro has the same core desktop dependencies.
Mint 21/22Ubuntu 22.04/24.04X11 / XWaylandCPU STT supported
Compatibility matrix
Supported, best-effort, and outside the current target.
This matrix is deliberately conservative. A successful OSS test on your actual desktop is more meaningful than a broad distro claim.
| Area | Supported | Best-effort / test carefully | Not currently documented |
|---|---|---|---|
| Distribution | Linux Mint 21/22; Ubuntu 22.04/24.04 | Closely related derivatives may work | Older than Ubuntu 22.04 / Mint 21 |
| Display session | X11; XWayland | Pure Wayland | Non-Linux desktops |
| Audio | PulseAudio; PipeWire compatibility | Unusual virtual devices and complex routing | Unsupported audio stacks |
| Text insertion | xdotool; clipboard workflows | ydotool; wl-paste; unusual toolkits | Protected fields that reject synthetic input |
| LLM | OpenAI-compatible chat endpoint | Provider-specific extensions | Endpoints without compatible chat streaming |
| Architecture | x86_64 Pro AppImage | OSS may be adaptable from source | Other Pro binaries unless separately published |
Pre-purchase test
A five-minute success is worth more than a compatibility slogan.
There is no Pro trial or post-delivery refund. Use the free edition as the required qualification path before checkout.
Confirm your session
Check whether the desktop is X11, XWayland, or pure Wayland.
Test input and output
Verify the microphone, normal speaker playback, dictation, and selected-text TTS.
Test target applications
Try the editor, browser, terminal, email client, and other windows you depend on.
Test Converse
Connect the actual LLM endpoint and model you intend to use; check first-response and streaming behavior.
Test an integration
When buying for Hermes or OpenClaw, run the integration’s real synthesis verification before purchasing.
Hardware guidance
No GPU required for speech recognition.
Model-server requirements are separate from AverVOX speech requirements. The LLM can run on the same machine, another LAN host, or a remote provider.
Microphone
A USB microphone or boom headset, stable distance, and low echo improve recognition and interruption reliability.
Model host
Choose hardware that fits the LLM. It can be more powerful—and physically separate—from the Linux desktop running speech.
Pure Wayland
Best-effort means test every critical interaction.
Wayland intentionally restricts global input and window access. Support depends on the compositor, desktop portal behavior, and the target application.
- Global hotkey registration
- Selected-text capture
- Clipboard read/write
- Synthetic text insertion
- Tray icon behavior
- Focus return after activation
A dash in this list does not mean the feature always fails; it means there is no single cross-compositor guarantee. X11 or XWayland is the supported path.
Compatibility questions
Qualify the environment before checkout.
Does Pro support more Linux desktops than OSS?
No broad compatibility advantage should be assumed. Pro bundles its runtime and adds features, but the core hotkey, audio, selection, and text-insertion path is shared.
Can I use a remote LLM from a low-power laptop?
Yes. Keep local speech on the laptop and configure an OpenAI-compatible endpoint on a stronger LAN machine or a remote provider.
Will an AppImage eliminate system dependencies?
Pro bundles its Python runtime, but desktop integration still depends on the supported display, audio, and input tools documented for the product.
Is Ubuntu 26.04 supported?
It is not included in the currently documented support matrix. Test OSS and wait for an explicit support statement before treating it as supported.
What should I include in a support report?
Distribution and version, session type, desktop environment, AverVOX version, audio stack, target app, insertion backend, exact reproduction steps, relevant logs, and whether the same action works from the CLI.
Start with the workflow
Test the exact workflow on the exact desktop.
AverVOX OSS is the compatibility check. Confirm every critical path before purchasing the perpetual Pro license.