Artificial intelligence

#Podcast: COMPRISE, the privacy-friendly, inclusive voice interface

Date:
Changed on 05/01/2022
Voice enabled apps are silently taking over our daily lives, mainly through our smartphones and digital assistants so that each one of us can easily order a meal or get an immediate answer on any subject that pops up in our mind. Easily? Not for everyone.
H2020 COMPRISE
© Inria / Photo D. Betzinger

 

What if the machine can’t understand your language or your accent? And how to guarantee your right to data-privacy? The COMPRISE project launched in late 2018 succeeded in creating such a privacy-friendly, more cost-effective and inclusive voice interface. We have here the honour of interviewing the project coordinator of COMPRISE, Emmanuel Vincent, from Inria Nancy – Grand Est (Multispeech team Inria-Loria), along with two of his colleagues from Inria Lille – Nord Europe, Nathalie Vauquier, the engineer in charge of Inria’s software development in COMPRISE, and Brij Srivastava, a young researcher who has joined Inria Startup Studio on the 1st of October 2021 to create a startup exploiting the results of the project.

What is the potential scientific and commercial impact of future software? What are the main challenges in developing voice recognition apps? Here are some of the topics the scientists are dealing with in this podcast.

visuel audio bloc par defaut
Titre du lecteur

Listen to the podcast

Fichier audio
Audio file