The purpose of this lecture is to expose the body of methods of analysis of musical and speech sound signals. The analysis, the modeling, the synthesis, the coding and speech recognition are treated. Measures, annexes but necessary for the methods described above, are also studied : the pitch, the voicing, the calculation of the MFCC, the calculation of the DTW, the detection of vocal activity... Some methods of spectral analysis must be studied in depth.

• Design a complete signal processing chain.
• Compare the performances of the various tools at our disposal for the analysis of complicated time series, in order to choose the one which will be best suited for this or that signal to be analyzed.
• Program in an interpreted computer language (matlab/octave/python/...).
• Mastering the basic and advanced principles of analog signal processing and digital signal processing.

Materials for the module is on edunao.
Course sequencing