In this thesis, a complete note onset estimation and pitch detection system ispresented, tuned for guitar music signals. The goal of this system is to rstestimate the point in time that a single note was played and then attributea frequency to it. This task is then extended to a sequence of played notes.Various onset and pitch estimation algorithms were tested and compared,taking into account both their eciency to complete the task, accurate tran-sciption of the signal, as well as their computational complexity timewise,thus achieving processing time less than or equal to the duration of the mu-sic signal.