The COMPRISE project (Cost-effective, Multilingual, Privacy-driven voice-enabled Services), a Research and Innovation Action (RIA) funded by the European Union's Horizon 2020 programme, had its kick-off meeting on 24th and 25th of January in Paris. The six consortium partners, Institut National de Recherche en Informatique et Automatique (Inria), Universitat de Saarlandes, Netfective Technology, Ascora, TILDE Y Rooter,, now have three years to create the next generation of voice interaction technology.
Voice-operated technologies and tools have multiplied in recent years: from Amazon's Alexa to voice-controlled radios or TVs, voice is rapidly replacing touch or text as the main means of interaction with modern devices. COMPRISE aims to support, and even lead on, this expansion by providing the tools and methodology to make voice interaction more secure, more cost-effective, and more inclusive for a variety of languages.
The scientists will rely on deep learning methodologies to improve speech-to-text and machine understanding of different languages. In addition, they aim to create a methodology that will protect the users' data, in order to ensure their privacy. Such innovations are expected to provide a strong basis for SMEs and European companies at large to design new services and expand.