Agathe Boudry
March 7, 2023
One of the integrated functions of the VIBE project is the Text-To-Speech and Speech-To-Text functionality. This feature allows the user to input written text and get the translated audio back or vice-versa.

STORY

For the implementation of the Text-To-Speech and Speech-To-Text function, Azure was used. Azure is responsible for managing and converting the input and output data. When the user talks or types in the VIBE application, the data gets sent to Azure which handles the conversion and sends the data back to the VIBE application. This data includes information for the digital human about audio, visemes and more. Thanks to the use of Azure it is possible to communicate in Dutch and English as well as select different male or female voices. Another upside to employing Azure is the easy integration with other dialog managers. This allowed for the integration of the speech and text system into the VIBE MMC use case.