Assignments

All assignments are due in your personal folder in the TECH 350 Box Folder by the assigned due date.

QUIZZES (15%)

Three quizzes (5% each) will be given during the first half of the course to test comprehension and retention of concepts related to 1) digital electronic music concepts (audio signal properties, signal quantization, additive/subtractive/modulation/granular synthesis), 2) DSP basics, block diagrams, digital delay lines, filters, and 3) different computational implementations of the Fourier transform. These quizzes will be a mixture of short answer, fill in the blank, and diagram-drawing.

MINI-ASSIGNMENTS (30%)

Ten mini-assignments (3% each) will be assigned during the first half of the course to reinforce the learning of concepts and tools including Max, Octave (alternative to Matlab), digital filter design, the DFT, STFT, and FFT, Sonic Visualizer, Max’s gen~, the JUCE framework, and Music Information Retrieval (MIR) techniques. These mini-assignments take the form of paper-and-pencil problem sets and coding tasks.

CLASS presentation (10%)

As part of the course that engages with uses of DSP within Music Information Retrieval (MIR) students will be asked to prepare and present a short (~5-minute) mini-lecture on a professor-assigned topic: low- and high-level features of music or digital audio techniques (see below). These presentations should include 1) a concise, clear introduction of the topic in plain english (3%), 2) a quantitative description of the topic through class-understandable mathematics and/or code (4%) and 3) a few examples of how the topic is engaged with in the context of DSP applications (for example: within an electronic artist’s musical practice, or a commercial digital audio tool/product) (3%).

Low-level Features:

  • Zero-Crossing Rate (ZCR) (rough pitch and brightness estimation)

  • f0 Estimation (fundamental pitch analysis)

  • Spectral descriptors (centroid, spread, kurtosis, flux, etc.)

  • Mel-Frequency Cepstral Coefficients (MFCCs)

  • Chromagram (pitch-class detection)

High-level Features:

  • Tempo estimation (100BPM or andante, for example)

  • Key estimation (Bb or Ab, for example)

  • Mood classification (happy, sad, or heroic, for example)

Digital Audio Techniques:

  • Karplus-Strong Synthesis (plucked string emulation technique)

  • Concatenative Synthesis (intelligent granular synthesis technique)

  • Phase-vocoding (flexible time- and frequency-shifting technique)

  • Linear Predictive Coding (LPC) (speech compression/processing method)

  • Deterministic-plus-Stochastic Model (DSM) (speech analysis/resynthesis method)

  • Dynamic Time Warping (DTW) (pre-processing for similarity measurement method)

FINAL project (45%)

The final project for this course will be a substantial individual project created over the second half of the course that builds off of, extends, dives deeper into, engages musically with, etc. the DSP concepts and tools engaged with in the first half of the course. At the beginning of the second half of the course brief in-class individual meetings will define your personal goals for the final project (5%). The final project could be a performable digital musical instrument (DMI), a sound generating or processing program, a polished VST plug-in, or a rigorous from-scratch implementation of a DSP technique, for example. The final project will be graded based on how well it accomplishes the goals set out in the individual meetings (20%). Additionally, a piece or corpus of music created with your final project (for example, a demonstration of its functionality, a musical composition created with it, etc.) should be submitted by the last day of class (10%). Lastly, a short technical paper that includes sections on the motivation for your final project, its design, its implementation (including uses of DSP techniques), and an evaluation of its success/applicability should be submitted by the end of reading period (10%).