This module has microphone and speakers (Left/Right, 1Watt total). This module can be used for voice control, voice assistance, text dictation, smart home and many more. Speech recognition and speech synthesis is processed on main Joker module using Neural Networks (NN). More detailed description can be found in this post.

TODO list:

• Far field microphone array for better voice capture (more distance, noise environments), like in Amazon Alexa;
• Low power voice activation mechanism. For example, FPGA based algorithm for continuous microphone monitoring for activation phrases (“Hey Joker” or another configured phrase).