Speech To Text
This is an STT engine developed on open source bases, based on Baidu's Deep Speech architecture, an open voice synthesis technology. In addition, its open architecture allows total transparency of its execution model and thus be easily implemented to the network of its users. All this allows a guarantee of privacy (privacy by design).
Our NLU brick allows a deep understanding of user interactions and related systems. Our NLU has 3 main responsibilities: to decrypt the users’ intentions, to detect their emotions and finally carry out associated actions. It is an elementary brick for the creation of virtual agents.
Our dialogue editor makes it easy to manage, update and extend the way a voice assistant's or bot's conversation should progress, in a graphical and ergonomic way.
The Dialogue Engine allows you to execute conversations that have been generated by the Dialogue Editor. This is the whole intelligent logic of the dialogue which is taken into account in this engine.
Text To Speech
This component takes the response from the wizard in text format and produces a voice expression that is returned to the user.