Install
Download the installer for Windows 10 or 11 and run it. The app, the voice engine and the virtual microphone are installed together.
Sign in with your account. The app registers this computer as a device on your account.
Documentation
How to set up ArVox and use every part of it.
Download the installer for Windows 10 or 11 and run it. The app, the voice engine and the virtual microphone are installed together.
Sign in with your account. The app registers this computer as a device on your account.
Pick your microphone and a voice model. Choose Local or Cloud.
Press Turn On Model. The button is available once the voice engine and the audio connector on this computer report that they are ready. If something is missing, the panel tells you which part.
In the app you call, stream or record with, choose our virtual microphone as the input.
Usage starts when the model is actually running and stops when you turn it off. Silence counts as usage while the model is on.
The free plan includes 15 minutes per day, reset at 00:00 UTC. Usage is tracked on our server from the engine session. If the connection drops, the session is closed at the last confirmed heartbeat.
Hear Myself plays the converted voice to your headphones so you can hear how you sound. Use headphones to avoid feedback.
The setting is saved to your account and applied by the desktop audio engine.
Give the model a name, choose its language and accent and upload recordings. You see file size, duration and upload progress.
Confirm that you have permission to use the voice, then submit. The training job status is reported by our training service: Uploaded, Queued, Processing, Completed or Failed.
Cloud Compute runs conversion on our GPU servers. You choose a daily allowance of GPU hours, which renews every day.
If the desktop benchmark shows your computer is too slow for real time, the app recommends Cloud.
On a team plan, invite people by email from Account. Each person uses one seat. Seats are separate from Cloud Compute.
Still stuck? Contact support.