Humanizing the Interface: Voice Activated Devices

Abstract

Speech recognition is the process of converting spoken words into machine readable format, suitable for processing. The project aims at controlling devices by recognizing voice commands. This is achieved by developing a stand-alone, speaker-dependent speech recognition system. The system takes in speech samples as input. It converts them to corresponding control signals for operating various devices using TI's DSP Starter Kit (TMS320C6713). Controlling devices by one's speech provides a user-friendly interface. With accurate recognition algorithms and noise cancellation techniques, the system is made capable of operating devices in various environments.