Music 254/CS 275b Spring 2021 Syllabus

Music 254/CS 275b "Computational Music Analysis "
Stanford University (Spring 2021).

This music information retrieval course (previously listed as "Music Query, Analysis, and Style Simulation") builds on the foundation of Music 253/CS 275A (Symbolic Musical Information). Using an open-source analysis platform (such as the Humdrum Toolkit (unix/bash shell), Humdrum Extras (C++), or Music21 (Python), or their preferred programming language/environment) students plan and design their own applications involving symbolic musical data.

The first four weeks cover basic areas of music analysis and retrieval and introduce basic principles of using the Humdrum Toolkit. Students may be assigned individual reading assignments pertinent to their research topic areas. The remaining six weeks are spent on individual project development, with weekly reports and discussions. Extensive help is available in the lab during hours allotted to class time. Students give verbal and written reports on their research at the end of the quarter. Some projects may be appropriate for continuation in independent-study modules.

For class Zoom link
Please contact esfield@stanford.edu or craig@ccrma.stanford.edu

Meeting times:		Tuesdays & Thursdays 12:30–1:50
Location:		Online
Instructors:		Eleanor Selfridge-Field (esfield@stanford.edu) Craig Sapp(craig@ccrma.stanford.edu)
Office Hours:		after class and by appointment on Zoom/Skype/Google Hangouts
Credits:		2–4
Grading:		25% class participation, 75% project Project proposal: 2+ pages, due Thurs. 23 April 2021 Project progress report: 4+ pages, Thurs. 14 May 2021 Project presentation in class: Thursday 4 June 2021 Project writeup: 10–20 pages Graduating on 14 June 2021: Due on 8 June 2021 10 p.m (exact deadline T.B.D.). Continuing at Stanford after 14 June 2021: Due on 11 June 2021 10 p.m (exact deadline T.B.D.).
Website:		music254.stanford.edu wiki.ccarh.org/wiki/Music 254 Overview of topics presented in Music 254.
Prerequisites:		Completion of Music 253/CS 275b, or demonstrated ability to use notation and sound software and associated requirements.
Textbook:		E. Selfridge-Field, ed., Beyond MIDI: The Handbook of Musical Codes (MIT Press, 1997). Available online by permission of the publisher.

Syllabus

Go to week: Harmony 1 | Melody 2 | Rhythm 3 | Harmony 4 | RegEx 5 | Programming 6 | 7 | 8

See also Music 253/CS 275a Syllabus

Week 1

Week	Dates	Topics
1	30 March, 1 April 2021	Lectures 1a. Symbolic Musical Analysis (Computational Music Analysis) (March 30) 1b. Melodic Search: Strategies and Format (April 1) Humdrum Lab 1 Humdrum data Harmony-related Humdrum tools Humdrum Harmony I hint: Harmonic intervals. tntype: twelve-tone chord prototypes. key: Krumhansl-Schmuckler key-finding algorithm. keycor: generalization of key command. sonority: triadic chord descriptions of sonorities. Performance analysis tool Sonic Visualiser Homework for new students For new students, enter this music in Verovio Humdrum Viewer. Try saving them as HTML scores from the File menu. Also useful, see this introduction to encoding music in Humdrum.

Week 2

Week	Dates	Topics
2	6 Apr 2021	Lectures: Approaches to melodic and other musical-similarity assessment Melodic Similarity Musical Similarity: More Perspectives Labs: Rhythm and score manipulation tools Rhythm Tools available in Humdrum timebase: create uniform duration for each Humdrum file data line. beat: label metric beats in Humdrum file. gettime: calculate performance time in seconds (according to tempo markings). thrux: expand repeats to performance sequence minrhy: identify minimum integral rhythmic unit in score assemble: merge parts into a single score partjoin: automation script to merge parts into a single score (preserves grace notes, uses minrhy and timebase) ditto/dittox: resolve null tokens hgrep: search for patterns in Humdrum file data.
2	8 Apr 2021	Data/Data entry Data/Data entry in Humdrum (PDF slides) KernScores: digital library of scores in the **kern format [1] introduction to KernScores [2] shortcuts to data in KernScores [3] browse collections in KernScores [4]: Josquin Research Project: digital library of early Renaissance polyphony xml2hum: MusicXML to Humdrum converter mid2hum: MIDI to Humdrum converter (works in many cases, but better to convert MIDI first to MusicXML then import). Regular Expressions (Text-based searching) Introduction to Regular Expressions

Week 3

Week	Dates	Topics
2	13 & 15 Apr 2021	Cope, Emmy, Emily Howell Lectures (Style Synthesis) Cope 1: Experiments in Musical Intelligence (Emmy) Cope 2: From Emmy to Emily Howell Melody Lab Humdrum Lab 2 Melody Tools available in Humdrum melody tools: mint: Melodic intervals scaletype Simple scale type categorizer tindex: Thema indexer. themax: Thema search. theloc: Thema localizer. context: melodic n-grams prange: Pitch range tool. support tools: kern: remove rhythms from note data. myank Yank measures from a full score. extract: extract spines from Humdrum file. extractx: Humdrum Extras version of extract hgrep: Humdrum-aware greep. autostem: automatic stemming of notes. unix: head: show only starting lines of text. grep: search text for generalized regular expressions. egrep: enhanced grep (adds additional wildcards). sed: Stream editor (search and replace). graphics (unix): pstopnm: a converter from PostScript to bitmaps. convert: Imagemagick tool for manipulating images. ps2pdf: convert PostScript into PDF files.

Week 4

Week	Dates	Topics
4	20 Apr 2021	Lectures: Complex Melodic Processes; Search Effectiveness and Modeling 4a. Complex Melodic Processes Humdrum parsing in C++ Harmony II Harmony II Key analysis Key-cycles: Hummel Op. 67 preludes Chopin Op. 28 preludes J.S. Bach Well-Tempered Clavier keycor: harmonic structure plots. mkeyscape: harmonic structure plots. Beethoven piano sonata keyscapes Mozart piano sonata keyscapes Lots of keyscapes transpose: Transpose score to specific key or by interval. Similarity with Normalized and Unnormalized Correlation
4	22 Apr 2021	Search Effectiveness 4b. Search Effectiveness and Modeling Humdrum & MIDI Humdrum to MIDI (PDF slides) Keyscapes Keyscapes

Week 5

Week	Dates	Topics
5	27 April 2021	Patterning and Coherence in Music 5a. Patterning and Coherence Guest lecture by Heinz Roggenkemper, "What is holding back optical music recognition?" Google slides Regular Expressions Regular Expressions (PDF slides)
5	29 April 2021	05B. Meter, Rhythm, and Expression prange and cint tools Performance analysis: Media:sonic-visualiser.pdf

Week 6

Week	Dates	Topics
6	4 May 2021	Counterpoint II Counterpoint II Rhythm and Meter Rhythm and Meter Programming for Humdrum files Humdrum Extras — C++ library for processing Humdrum files Ratioscores: A text-based system for just intervals, Julie Zhu and Craig Sapp
6	6 May 2021	Sonority distributions by bass-line scale degrees Corelli & Rule-of-the-octave data selection (PDF slides)

Week 7

Week	Dates	Topics
7	11 May 2021	Recent developments in MIR Recent Work in Symbolic MIR
7	13 May 2021	Recent Work in Harmonic Analysis Harmonic Models Keyscapes

Week 8

Week	Dates	Topics
8	18 May 2021	Project development
8	20 May 2021	Project development Performance style analysis

Week 9

Week	Dates	Topics
9a	25 & 27 May 2021	Counterpoint Generative approaches to counterpoint MEI MEI (Music Encoding Initiative) Release 2.0

Week 10

Week	Dates	Topics
10	4 June 2020	Project presentations in class. Submit to Craig and Eleanor before class starts.
11	7 June 2020	Project write-ups for students taking a degree in June 2021.
12	11 June 2021	Project write-ups for continuing students continuing at Stanford (Deadline T.B.D.).

Other topics/resources

• • Support tools:
    • assemble: merge separate spines into single data stream.
    • serialize: convert Humdrum data into single spine.
    • humcat : concatenate multiple scores together (download data from web).
    • ridx: remove various components of Humdrum file structure
    • ditto: resolve meaning of null tokens.
    • humsplit: split segmented Humdrum file streams into separate files.
  • Unix tools:
    • command line
    • list of command-line programs
    • sort: sort lines of data alphabetically or numerically.
    • uniq: Remove/count successively repeated lines of text.
• Pitchmix and Markov analysis of composite sonorities

General Policies and University Rules

General policies and university rules:

1. Delivery times:
   1. Assignments: by the start of the class for which they are due.
   2. Project writeup: by 11 p.m. 7 or 11 June 2021 by 11p.
2. Honor code: We will act and expect you to act according to the Stanford Honor Code.
3. Students with disabilities: Students who may need an academic accommodation based on the impact of a disability must initiate the request with the Student Disability Resource Center (SDRC) located within the Office of Accessible Education (OAE). SDRC staff will evaluate the request with required documentation, recommend reasonable accommodations, and prepare an Accommodation Letter for faculty dated in the current quarter in which the request is being made. Students should contact the SDRC as soon as possible since timely notice is needed to coordinate accommodations. The OAE is located at 563 Salvatierra Walk (phone 723-1066).
4. Students in distress: Stanford offers an abundance of resources for students dealing with problems of a personal nature.
   1. Confidential support: (650) 736-6933
   2. Crisis hotline: (650) 725-9955
   3. Title IX questions: (650) 497-4955