Table of Contents
1. System Prerequisites
Before initiating real-time speech translation during live online meetings, ensure your workstation meets the following hardware and software requirements:
- OS: Windows 10 or Windows 11 (64-bit).
- RAM: 8 GB minimum (16 GB recommended for seamless multitasking during video calls).
- Audio Interface: Standard built-in soundcard supporting Windows Audio Session API (WASAPI).
- Storage: ~850 MB free disk space for local AI model files.
2. Installation & Language Pack Download
Download the latest standalone executable setup file from Voxxwire's official download page. Run Voxxwire-Setup.exe and follow the wizard.
Upon launching Voxxwire for the first time:
- Select your Source Language (e.g. Japanese, German, Spanish) and Target Language (e.g. English).
- Choose your preferred Whisper ASR model size (e.g.
baseorsmall). - Click Download Language Pack. The app will fetch the corresponding offline translation pair once. Subsequent launches require zero internet access.
Advertisement
3. Configuring Audio Devices (Mic & System Output)
Voxxwire uses dual-channel audio capture. Channel 1 captures your local microphone (your voice), while Channel 2 captures WASAPI system audio loopback (the remote speakers in Zoom or Teams).
# Device Selection Matrix
Local Microphone --> Input Channel 1 (Your Speech) --> Whisper ASR --> Translation
System Loopback --> Input Channel 2 (Remote Speech) --> Whisper ASR --> Translation
In the Voxxwire main control panel:
- Set Microphone Input: Choose your default physical mic (e.g. Realtek High Definition Audio or USB Headset).
- Set System Audio Input: Select Default System Output (WASAPI Loopback).
4. Connecting Voxxwire to Zoom & Microsoft Teams
Because Voxxwire captures system audio at the OS driver layer, no Zoom app plugins or custom Teams permissions are required!
- Start your meeting in Zoom, Microsoft Teams, or Google Meet as usual.
- Position the Voxxwire window alongside or overlaying your meeting window.
- Click Start Listening in Voxxwire.
- As participants speak in the foreign language, real-time dual-line subtitles appear instantly on screen showing the original transcript and English translation side-by-side.
5. Troubleshooting Audio Loopback & Performance
If subtitles do not appear when remote participants speak:
- Verify that your headset or speakers are set as the Default Communications Device in Windows Sound Control Panel.
- Ensure WASAPI loopback isn't muted in Windows Volume Mixer.
- If using high-CPU video effects in Zoom, lower the Whisper model setting in Voxxwire from
smalltobaseortinyto reduce processing load.