Voxnara

Preparing your workspace

Prepare

Download and choose a local model

Install a local recognition model before starting microphone, desktop, or file transcription.

Updated Sep 21, 2026

Speech to Text requires an installed local recognition model before microphone, desktop, or file transcription can begin.

Follow these steps

  1. 01

    Open Provider connections and select the Speech to Text tab.

  2. 02

    Review each model’s size and expected performance, then download the model suited to your device.

  3. 03

    Wait for the download and verification to complete before opening Speech to Text.

  4. 04

    Select the installed model and the language spoken in the recording.