Real-time users voice localization (subtitles)

[thedrhax14]

Hello Dolby Team,

Is there a way to implement real-time translation of users' speech with subtitles? For example, lets say each remote user is presented in their own window in grid layout where each window has their name, avatar and video feed. Then local user enables subtitles and now for each remote user there is text at the bottom of their window like a youtube video the text appears in real-time in a language that local user have chosen.

Bests,
thedrhax14.

[j12y]

Hello, let me see if I can help.

We've recently released a solution Real-time Media Extensions which provides an integration point for implementing solutions such as live transcription. This works in conjunction with our Communications APIs.

You can find some background information here:
https://api-references.dolby.io/real-time-media-extensions/docs/introduction

The GitHub repository here:
https://github.com/DolbyIO/real-time-media-extensions

We don't currently provide the translation or transcription services directly, they must be setup as an extension and it is up to the front-end developer to display the returned transcription / translation text result in the UX. With Real-time Media Extensions we provide the hook for making the connection from a Dolby.io Video Conference to receive and forward the audio stream to the appropriate transcription services.