Caption video streams with multi-track audio – CaptionHub

This guide is intended for customers who need to caption multiple audio tracks within the same video stream.

A common example is a live event where the primary audio is being translated by an on-site interpreter. Each language audio feed can be captioned independently and then made available within the same stream.

It’s only available with the Player plugin output.

How it works

Multi-track captioning works by creating multiple linked Timbra projects, with each project configured to transcribe a different audio track from the same HLS stream.

One project acts as the parent project, while the additional projects act as sub-projects for the other audio tracks. Captions generated by the sub-projects are automatically surfaced within the parent project.

Setting up multi-track audio

Once multi-track audio has been enabled, you can create additional audio track projects from the parent project.

Open the project menu

Select Duplicate for multi-track

Configure the new project to use the appropriate audio track

Change the source language to the language of the audio track

This creates a linked sub-project. Any captions generated by the sub-project are automatically pulled into the parent project.

Usage notes

Each project must be started individually

Each project must be stopped individually

Ensure that only one language is configured as Player default

All linked projects should be configured before the event begins to ensure a consistent viewer experience