XMOS voice solutions make voice interfacing ubiquitous across our home, office and leisure lives. Just as we talk to people we have never met before and chat with family and friends, so we will use voice to communicate with the 100s of machines we will interact with in future.
Its controllers integrate all the features of a voice interface in a single silicon device that provides the fastest and most cost effective way to deliver voice-enabled products to market.
- voice capture from large numbers of microphones and aggregate those voice inputs;
- voice activity detection to wake always-on devices from low power sleep modes;
- sophisticated digital signal processing on the captured signal to address the problems of distance, isolation, the removal of noise, and the establishment of direction of voice capture;
- transfer of the processed voice samples to automatic speech recognition systems hosted in the cloud or simpler systems hosted in the piece of embedded equipment itself;
- response to instructions from the embedded system or from the cloud-based system and interaction with the environment around us.