Localizing and Identifying Audio or Voice Sources with a MEMS Microphone Array

In the last decade, interaction between humans and electronic device by means of spoken language and audio signals has grown more and more important in everyday’s life. Different capabilities have to be exploited by the device itself to achieve the best performances in this field, like audio localization, automatic speech recognition, speaker identification, and so on. In this paper we focus on both the algorithms adopted to gain these results and the hardware architecture that will enable the embedding of the whole application: the MEMS microphone array.