Abstract
In the past few years, there has been continuing increase in the number and range of products and services that incorporate speech technology. This increase in applications is partly due to the advances made in field of speech synthesis and recognition; and partly due to the tremendous progress made in high performance computing (HPC) industry. Scientific innovations that were just imagination before couple of decades are becoming reality now because of progress made in audio signal processing field. Voice-activated technology has entered our lives in professional, personal as well as recreational applications. Touch sensitive interface in desktops, cell phones, music i-pods etc are nearly inventions of the past; as callers needs simply to speak the name of the menu or option they desire into the phone. Nearly every cell phone in the market comes equipped with voice-activated dialing, allowing users just to speak the name of the person they wish to call instead of browsing through a list of contacts. Moreover, in this technologically advanced world, equipped with ATM debit cards, credit cards, driver’s licenses, online banking facilities and similar other types of identity proofs could be hampered. Although most of the technological innovations are made for the welfare of the society, some destructive minds tend to take its undue advantage. Therefore, there are always higher chances that an individual’s identity is stolen without his or her knowledge. The physical identifications can be lost or stolen and even the signatures can be counterfeit. The purpose of this project is to make study of Linear Predictive Coding algorithm that is currently used in many control and security systems. This project is focused on studying and implementing Linear Predictive Coding (LPC) algorithm in MATLAB for audio feature extraction from a given speech signal. Along with that, digital filter design is also implemented for audio feature extraction from the given speech/sample input signal. Towards the end of the project, I focused on designing the input circuitry (that consisted of microphone and pre-amplifier stage) for next phase of the project.