Audio and Speech Processing

Signal-processing and machine-learning techniques for speech, music and other audio signals is the topic. The basics for speech production and hearing are introduced. PDEs and simplified vocal-tract models are derived for speech and acoustic propagation models are described. LPC, DFT/cepstral audio analysis methods are discussed as well as the modern method for speech synthesis. Basic dynamic programming and hidden Markov modeling are introduced. Microphone-array methods are presented. Coding methods for speech and music are included. Real-time issues are considered. A project and presentation are important in grading. Offered every other year.