Fixed installation: Dante Streaming Setup to LiveVoice

Overview

This guide explains how to stream a translator's audio feed via Dante Virtual Soundcard (DVS) to LiveVoice. The goal is to make the translation accessible to listeners through the LiveVoice app.


Thanks to ICF Zürich that provided this practical guide to share how they have done their setup!

  1. Requirements

Hardware:

    • Headset: Sennheiser HME 27
    • Console: Studio Technologies Inc. Model 204 Announcers Console

    • Computer: Intel NUC running Windows

      Software:

    • Dante Virtual Soundcard (DVS): To receive audio
    • FFmpeg: To stream the audio feed via RTP
    • LiveVoice: Online platform for audio streaming


  1. Setup Steps

Connect Hardware:

    • Connect the headset to the announcers console.

    • Link the console to the Intel NUC or macOS device via the Dante network.

      Install Software:

    • Install Dante Virtual Soundcard (DVS) and configure the appropriate channels (e.g., Receive 1-2 for English).

    • Download FFmpeg from the official website and extract it to your device.

      Set Up the LiveVoice Channel:

    • Log in to LiveVoice and enable the “Import Audio” option on the desired channel.
    • Select RTP as the protocol and OPUS as the audio format.
    • Copy the stream URL and port provided by LiveVoice into the FFmpeg script.

Streaming Script for Windows

For Windows, use the following commands:

  • List Available Audio Devices:
ffmpeg -list_devices true -f dshow -i dummy
  • Streaming Script: Save the following script as a .bat file, e.g., EnglishStream.bat :
@echo off
TITLE Livevoice English Stream from DVS Receive 1-2
ffmpeg -f dshow -i audio="DVS Receive  1-2 (Dante Virtual Soundcard)" -acodec libopus -bufsize 500k -ab 192k -ac 2 -f rtp rtp://<LiveVoice-URL>:<Port>
exit

Replace <LiveVoice-URL> and <Port> with the values provided by LiveVoice.

Streaming Script for Mac

For Mac, use the following commands:

  • List Available Audio Devices:
./ffmpeg -f avfoundation -list_devices true -i ""
  • Streaming Script: Save the following as a shell script file (e.g., EnglishStream.sh ):
#!/bin/bash
ffmpeg -f avfoundation -i ":0" -acodec libopus -bufsize 500k -ab 192k -ac 2 -f rtp rtp://<LiveVoice-URL>:<Port>
  • Replace <LiveVoice-URL> and <Port> with the values provided by LiveVoice. Make the script executable by running:
chmod +x EnglishStream.sh

Configure Autostart

  • Windows: Place the .bat file in the Windows startup folder:
C:\Users\<Username>\AppData\Roaming\Microsoft\Windows\Start Menu\Programs\Startup
  • Mac: Add the shell script to a login item using macOS System Preferences or a third-party automation tool.

Perform a Soundcheck

Before the event, test the setup by streaming an MP3 file:

  • Windows:
ffmpeg -re -i test.mp3 -acodec libopus -bufsize 500k -ab 192k -ac 2 -f rtp rtp://<LiveVoice-URL>:<Port>
  • Mac:
./ffmpeg -re -i test.mp3 -acodec libopus -bufsize 500k -ab 192k -ac 2 -f rtp rtp://<LiveVoice-URL>:<Port>

Verify the connection and audio quality in the LiveVoice app.

Troubleshooting

  • No Audio Input Detected: Ensure the correct channel (e.g., “DVS Receive 1-2”) is selected in FFmpeg.
  • RTP Stream Not Reachable: Check network connectivity and firewall settings.
  • Low Audio Quality: Adjust the bitrate (-ab ) and buffer size (-bufsize ) in the FFmpeg script.