Difference between revisions of "MIR workshop 2012"

From CCRMA Wiki
Jump to: navigation, search
(Day 4: Pitch, Chroma, More Classification)
 
(67 intermediate revisions by 3 users not shown)
Line 2: Line 2:
  
 
== Logistics ==
 
== Logistics ==
Workshop Title: '''"Intelligent Audio Systems: Foundations and Applications of Music Information Retrieval"
+
Workshop Title: '''Intelligent Audio Systems: Foundations and Applications of Music Information Retrieval'''
'''
+
* Monday, June 25, through Friday, June 29, 2012. 9:30 AM (not 9 AM) to 5 PM every day.
* 9-5 PM.
+
* Location: The Knoll, CCRMA, Stanford University. http://goo.gl/maps/nNKx
 
* Instructors:  
 
* Instructors:  
- Jay LeBoeuf,  
+
** Jay LeBoeuf, [http://www.izotope.com iZotope, Inc.], [http://www.imagine-research.com Imagine Research, Inc.]
  - Steve Tjoa
+
** Steve Tjoa, [http://www.izotope.com iZotope, Inc.]
  - Leigh Smith,   
+
** Leigh Smith,  [http://www.izotope.com iZotope, Inc.]
 +
 
 +
2013
 +
[[MIR_workshop_2013]]
  
 
* Participants:
 
* Participants:
Line 24: Line 27:
 
'''Workshop structure:''' The workshop will consist of half-day lectures, half-day supervised lab sessions, demonstrations, and discussions. Labs will allow students to design basic ground-up "intelligent audio systems", leveraging existing MIR toolboxes, programming environments, and applications. Labs will include creation and evaluation of basic instrument recognition, transcription, and real-time audio analysis systems.
 
'''Workshop structure:''' The workshop will consist of half-day lectures, half-day supervised lab sessions, demonstrations, and discussions. Labs will allow students to design basic ground-up "intelligent audio systems", leveraging existing MIR toolboxes, programming environments, and applications. Labs will include creation and evaluation of basic instrument recognition, transcription, and real-time audio analysis systems.
  
== Lectures & Labs ==
+
== Schedule: Lectures & Labs ==
<br><u>Day 1:</u> [https://ccrma.stanford.edu/workshops/mir2011/CCRMA_2011_day1.pdf Lecture 1 Slides]
+
 
* '''Presenters: Jay LeBoeuf & Rebecca Fiebrink'''
+
=== Day 1: Introduction to MIR, Signal Analysis and Feature Extraction ===
 +
Presenters: Jay LeBoeuf, Leigh Smith
 +
 
 +
<br><u>Day 1: Part 1</u> [http://ccrma.stanford.edu/workshops/mir2012/CCRMA2012day1.pdf Lecture 1 Part 1 Slides]
 +
* Introductions 
 
* CCRMA Introduction - (Carr/Sasha).  CCRMA Tour.
 
* CCRMA Introduction - (Carr/Sasha).  CCRMA Tour.
 
* Introduction to MIR (What is MIR? Why are people interested? Commercial Applications of MIR)   
 
* Introduction to MIR (What is MIR? Why are people interested? Commercial Applications of MIR)   
* A brief history of MIR
 
** See also http://www.ismir.net/texts/Byrd02.html
 
 
* Overview of a basic MIR system architecture     
 
* Overview of a basic MIR system architecture     
 
* Timing and Segmentation: Frames, Onsets       
 
* Timing and Segmentation: Frames, Onsets       
 
* Features: ZCR, Spectral moments; Scaling of feature data   
 
* Features: ZCR, Spectral moments; Scaling of feature data   
 +
* Demo: Using simple heuristics and thresholds (i.e. "Why do we need machine learning?")
 
* Classification: Instance-based classifiers (k-NN)   
 
* Classification: Instance-based classifiers (k-NN)   
* Information Retrieval Basics
+
* Information Retrieval Basics (Part 1)
 
** Classifier evaluation (Cross-validation, training and test sets)  
 
** Classifier evaluation (Cross-validation, training and test sets)  
** IR Evaluation Metrics (precision, recall, f-measure, AROC,...)
 
*** [http://ccrma.stanford.edu/workshops/mir2009/references/recall_precision.pdf Recall-Precision]
 
*** [http://ccrma.stanford.edu/workshops/mir2009/references/ROCintro.pdf ROC Analysis]
 
  
* Application: Instrument recognition and drum transcription / Using simple heuristics and thresholds (i.e. "Why do we need machine learning?")
+
<br><u>Day 1: Part 2</u> [http://ccrma.stanford.edu/workshops/mir2012/CCRMA_MIR2012_FeatureExtraction.pdf Lecture 2 Slides]
<br><u>Lab 1:</u> <br>
+
* [https://ccrma.stanford.edu/workshops/mir2011/Lab_1_2011.pdf Lab 1 - Basic Feature Extraction and Classification] <br>
+
* [http://ccrma.stanford.edu/workshops/mir2011/weka_lab1.pdf Getting started with Weka]
+
* [https://ccrma.stanford.edu/workshops/mir2011/Wekinator_lab_2011.pdf Wekinator Lab]
+
* Overview of Weka & the Wekinator
+
** [http://www.cs.waikato.ac.nz/ml/weka/ Weka home]
+
** [http://code.google.com/p/wekinator/ Wekinator on Google code] and [http://wiki.cs.princeton.edu/index.php/ChucK/Wekinator/Instructions instructions]
+
Students who need a personal tutorial of Matlab or audio signal processing will split off and received small group assistance to bring them up to speed.
+
* Background for students needing a refresher:
+
** [http://ccrma.stanford.edu/workshops/mir2009/juans_lecture/2_fft.pdf Fundamentals of Digital Audio Signal Processing (lecture slides from Juan Bello)]
+
** [http://ccrma.stanford.edu/workshops/mir2009/Lab0/lab0.html Fundamentals of Matlab]
+
** [http://ccrma.stanford.edu/workshops/mir2009/Lab1/lab1.html Fundamentals of Digital Audio Signal Processing (FFT, STFT, Windowing, Zero-padding, 2-D Time-frequency representation)]
+
  
* REMINDER: Save all your work, because you may want to build on it in subsequent labs.
 
 
<br><u>Day 2:</u> [https://ccrma.stanford.edu/workshops/mir2011/CCRMA_2011_day2.pdf Lecture 2 Slides]
 
[http://ccrma.stanford.edu/workshops/mir2011/BeatReferences.pdf A list of beat tracking references cited]
 
* '''Presenters: Leigh Smith & Stephen Pope'''
 
 
* Overview: Signal Analysis and Feature Extraction for MIR Applications (Historical: http://quod.lib.umich.edu/cgi/p/pod/dod-idx?c=icmc;idno=bbp2372.1999.356)
 
* Overview: Signal Analysis and Feature Extraction for MIR Applications (Historical: http://quod.lib.umich.edu/cgi/p/pod/dod-idx?c=icmc;idno=bbp2372.1999.356)
 
* MIR Application Design
 
* MIR Application Design
Line 82: Line 68:
 
* Other Feature domains
 
* Other Feature domains
 
** Wavelets, LPC
 
** Wavelets, LPC
 +
 +
 +
* Application: Instrument recognition and drum transcription / Using simple heuristics and thresholds (i.e. "Why do we need machine learning?")
 +
<br><u>Lab 1:</u> <br>
 +
* [https://ccrma.stanford.edu/workshops/mir2011/Lab_1_2012.pdf Lab 1 - Basic Feature Extraction and Classification] <br>
 +
Students who need a personal tutorial of Matlab or audio signal processing will split off and received small group assistance to bring them up to speed.
 +
* Background for students needing a refresher:
 +
** [http://ccrma.stanford.edu/workshops/mir2009/juans_lecture/2_fft.pdf Fundamentals of Digital Audio Signal Processing (lecture slides from Juan Bello)]
 +
** [http://ccrma.stanford.edu/workshops/mir2009/Lab0/lab0.html Fundamentals of Matlab]
 +
** [http://ccrma.stanford.edu/workshops/mir2009/Lab1/lab1.html Fundamentals of Digital Audio Signal Processing (FFT, STFT, Windowing, Zero-padding, 2-D Time-frequency representation)]
 +
 +
* REMINDER: Save all your work, because you may want to build on it in subsequent labs.
 +
 +
=== Day 2: Beat-finding and Rhythm Analysis ===
 +
Presenter: Leigh Smith
 +
 +
[http://ccrma.stanford.edu/workshops/mir2012/CCRMA_MIR2012_Beat.pdf Lecture 3 Slides]
 +
[http://ccrma.stanford.edu/workshops/mir2011/BeatReferences.pdf A list of beat tracking references cited]
 +
 
* Onset-detection: Many Techniques
 
* Onset-detection: Many Techniques
 
** Time-domain differences
 
** Time-domain differences
Line 100: Line 105:
 
** Fluctuation Patterns
 
** Fluctuation Patterns
 
** Joint estimation of downbeat and chord change
 
** Joint estimation of downbeat and chord change
* Applications
 
** Song clustering based on a variety of feature vectors
 
** PCA of feature spaces using Weka
 
 
<br><u>Lab 2:</u>  
 
<br><u>Lab 2:</u>  
* Feature extraction and flexible feature vectors in MATLAB, Marsyas, Aubio, libExtract
+
* [http://ccrma.stanford.edu/workshops/mir2012/FeatureDetection_lab2_2012.pdf Feature extraction and cross-validation in MATLAB]
* MATLAB/Weka code for sound clustering with a flexible feature vector
+
* C++ API examples Marsyas, Aubio, libExtract - pre-built examples to read and customize
+
* Extract CAL 500 per-song features to .mat or .csv using features from today.  This will be used on lab for Friday. Copy it from the folder ccrma-gate.stanford.edu:/usr/ccrma/workshops/mir2011/cal500.tar (beware it's a 2Gb .tar file!) or grab the AIFF versions from ccrma-gate.stanford.edu:/usr/ccrma/workshops/mir2011/cal500_aiffs.tar (that's 16 GB)
+
 
* Down-loads
 
* Down-loads
 
**  [https://ccrma.stanford.edu/workshops/mir2011/MAT240F-Reader.zip UCSB MAT 240F Reader]
 
**  [https://ccrma.stanford.edu/workshops/mir2011/MAT240F-Reader.zip UCSB MAT 240F Reader]
 
**  [https://ccrma.stanford.edu/workshops/mir2011/MAT240F-Code.zip UCSB MAT 240F Code]
 
**  [https://ccrma.stanford.edu/workshops/mir2011/MAT240F-Code.zip UCSB MAT 240F Code]
 
**  [https://ccrma.stanford.edu/workshops/mir2011/MAT240F-Sounds.zip UCSB MAT 240F Sounds]
 
**  [https://ccrma.stanford.edu/workshops/mir2011/MAT240F-Sounds.zip UCSB MAT 240F Sounds]
**  [https://ccrma.stanford.edu/workshops/mir2011/ODF.zip Onset Detection Function example code in Octave/Matlab]
+
**  [https://ccrma.stanford.edu/workshops/mir2012/ODF.zip Onset Detection Function example code in Octave/Matlab]
* Notes on c-API configuration
+
 
** FFTW
+
=== Day 3: Music Information Retrieval in Polyphonic Mixtures ===
./configure --help<br>
+
Presenter: Steve Tjoa
./configure --enable-float
+
 
** libSndFile
+
'''Lecture/Lab 3'''
./configure --disable-external-libs --disable-sqlite
+
* [http://ccrma.stanford.edu/workshops/mir2012/tjoa20120627ccrma.pdf Lecture and Lab 3 Slides, Steve Tjoa, 2012]
** CAL500 decoding
+
 
for i in *.mp3; do echo $i; afconvert -d BEI16@44100 -f AIFF "$i"; done
+
* Music Transcription and Source Separation
<br><u>Day 3</u>
+
* Nonnegative Matrix Factorization
* '''Presenters: Stephen Pope & Steve Tjoa'''
+
* Sparse Coding
* [http://up.stevetjoa.com/tjoa20110629ccrma.pdf Lecture and Lab 3 Slides by Steve Tjoa]
+
* Locality Sensitive Hashing
* [https://ccrma.stanford.edu/workshops/mir2011/CCRMA_2011_day3.pdf Lecture 3 Slides]
+
 
* Overview: 2nd-Stage Processing and Post-processing in MIR Applications
+
=== Day 4: Pitch, Chroma, More Classification ===
* 2nd-Stage Processing
+
 
** Thresholds and Data Pruning
+
Presenters: Steve Tjoa, Oscar Celma (Gracenote)
** Perceptual Mapping
+
 
** Data Reduction:  Averaging, GMMs, Running Averages
+
'''Lecture 4'''
** Feature-data-smoothing: de-spiking, sticky values, filter, etc.
+
* [http://ccrma.stanford.edu/workshops/mir2011/ccrma_2011_pitch_reps.pdf Pitch Representation Slides (.pdf), George Tzanetakis]
* Segmentation of music and non-musical audio
+
** Segmentation based on islands of similar features
+
** Segmentation based on regular difference peaks
+
** Segmentation based on labeling
+
* Post-processing: What are we doing?
+
** Storing Feature Data: SQL, JSON, XML, etc.
+
** Classification/Clustering/Transcription/Labeling
+
* Classification: KNN vs SVM training and testing
+
** SVM tools and APIs
+
* Clustering vs Classification: Tree-based systems
+
* Audio Transcription: Onsets and per-onset features
+
* Other applications: source separation, similarity match, search, etc.
+
* Classification/estimation in the presence of polyphony
+
** Try basic approach on a musical mixture.
+
** How well does it perform?
+
** What do we do to improve its performance? ICA, NMF, K-SVD.
+
** Matrix representations of data: spectrogram, chromagram, timbregram, etc.
+
** Methods to improve NMF/K-SVD under heavy harmonic overlap
+
* Applications
+
** Feature vector pruning
+
** Segmentation examples
+
** SVMs for classification
+
** Multipitch estimation, source separation, denoising
+
<br><u>Lab 3:</u>
+
* 2nd-Stage Processing
+
* SVM tools
+
* Classification examples
+
<p>
+
* If you finish early, see the "bonus labs" section below.
+
  
<br><u>Day 4:</u>
 
* '''Presenters: George Tzanetakis'''
 
* [http://ccrma.stanford.edu/workshops/mir2011/ccrma_2011_pitch_reps.pdf Pitch Representation Slides (.pdf)]
 
* [http://ccrma.stanford.edu/workshops/mir2011/ccrma_2011_pitch_reps.pptx Pitch Representation Slides (.pptx)]
 
 
* Features:  
 
* Features:  
 
** Monophonic Pitch Detection  
 
** Monophonic Pitch Detection  
Line 176: Line 142:
 
** Audio-Score Alignment  
 
** Audio-Score Alignment  
 
** Cover Song Detection  
 
** Cover Song Detection  
** Analysis of religious cantillation (Computational Ethnomusicology)
 
 
** Query-by-humming  
 
** Query-by-humming  
 
** Music Transcription  
 
** Music Transcription  
* Tools
 
** Marsyas
 
** Python/NumPy/Matplotlib
 
<br><u>Lab 4: </u>
 
* Marsyas compilation
 
** Instructions for CCRMA Machines [http://ccrma.stanford.edu/workshops/mir2011/marsyas_ccrma2011.pdf marsyas_ccrma2011.pdf]
 
*** SKT: If you get an error about Python.h, install the package python2.7-dev (for version 2.7).
 
* Marsyas tour
 
* Plotting and prototyping using the Marsyas Python bindings
 
* Writing some C++ Marsyas code
 
* DTW in Matlab [http://labrosa.ee.columbia.edu/matlab/dtw/ Dan Ellis DTW Matlab example]
 
 
 
<br><u>Day 5:</u>
 
* '''Presenters: Douglas Eck'''
 
* [http://ccrma.stanford.edu/workshops/mir2011/CCRMA_2011_day5.pdf Day 5 Slides (.pdf)]
 
 
* Music Recomendation
 
* Music Recomendation
 
** Overview of music recommendation. What's hard about it.
 
** Overview of music recommendation. What's hard about it.
 
** Some statistics and observations about the music industry and the need for recomendation.
 
** Some statistics and observations about the music industry and the need for recomendation.
 
** Point-Counterpoint: Should we bother with content-based analysis.
 
** Point-Counterpoint: Should we bother with content-based analysis.
*  Autotagging
+
 
** Features for autotagging (some of this will be review, given days 1 through 4.)
+
'''Lab 4'''
** Demos of clustering for different types of acoustic features.
+
* [http://ccrma.stanford.edu/workshops/mir2012/2012-ClusterLab.pdf K-Means]
** Training Data.
+
* [http://ccrma.stanford.edu/workshops/mir2012/Lab5-SVMs.pdf SVM]
** Classifiers (focus on AdaBoost)  
+
 
** Feature selection.
+
=== Day 5: Michael Mandel, Jay LeBoeuf, Steve Tjoa, Leigh Smith ===
** Evaluation with lots of examples.
+
Guest Lecture: Autotagging, Michael Mandel
 +
* [http://majorminer.org/info/intro Major Miner Game]
 +
* [http://musicallyintelligent.com/ Musically Intelligent Machines]
 +
 
 +
* [https://ccrma.stanford.edu/workshops/mir2012/CCRMA%202012%20day1%20v5.pdf Day 5 Slides (.pdf)]
 +
 
 +
References:
 +
** IR Evaluation Metrics (precision, recall, f-measure, AROC,...)
 +
*** [http://ccrma.stanford.edu/workshops/mir2009/references/recall_precision.pdf Recall-Precision]
 +
*** [http://ccrma.stanford.edu/workshops/mir2009/references/ROCintro.pdf ROC Analysis]
 +
 
 
* Time permitting:
 
* Time permitting:
 
** Advanced features  
 
** Advanced features  
Line 211: Line 170:
 
** Using musical structure.
 
** Using musical structure.
  
<br><u>Lab 5</u>
+
'''Lab 5'''
  
See [[MIR_workshop_2011_day5_lab]] for a full description. Here is a summary:
+
Below are some links to functions and matlab code for key estimation, chord recognition, and GMMs:  
 
+
** [http://ccrma.stanford.edu/workshops/mir2009/Lab3/lab3.tgz Lab - download lab3.tgz]
* The basics (some Python code available to help).
+
** [http://ccrma.stanford.edu/workshops/mir2009/Lab3/lab3.html Lab - Key estimation, chord recognition]
** Calculate acoustic features on CAL500 dataset (students should have already done this.)
+
** [http://ccrma.stanford.edu/workshops/mir2010/Lab4_2010.pdf GMM Lab]
** Read in user tag annotations from same dataset provided by UCSD.
+
** Build similarity matrix based on word vectors derived from these annotations.
+
** Query similarity matrix with a track to get top hits based on cosine distance.
+
** Build second similarity matrix using acoustic features.  
+
** Query this similarity matrix with track to get top hits based on cosine distance.  
+
  
* Extra (I didn't write code for this, but can help students find examples).
 
** Query the EchoNest for additional acoustic features and compare to yours.
 
** Use the CAL500 user annotations as ground truth and evaluate your audio features (ROC curve or some precision measure).
 
** Compare a 2D visualization of acoustic features versus UCSD user annotations.
 
 
<br>
 
<br>
 
<br><u>Bonus Lab material</u>
 
<br><u>Bonus Lab material</u>
Line 236: Line 186:
 
** [http://ccrma.stanford.edu/workshops/mir2009/Lab3/lab3.tgz Lab - download lab3.tgz]
 
** [http://ccrma.stanford.edu/workshops/mir2009/Lab3/lab3.tgz Lab - download lab3.tgz]
 
** [http://ccrma.stanford.edu/workshops/mir2009/Lab3/lab3.html Lab - Key estimation, chord recognition]
 
** [http://ccrma.stanford.edu/workshops/mir2009/Lab3/lab3.html Lab - Key estimation, chord recognition]
 +
 +
* [http://ccrma.stanford.edu/workshops/mir2011/weka_lab1.pdf Getting started with Weka]
 +
* [https://ccrma.stanford.edu/workshops/mir2011/Wekinator_lab_2011.pdf Wekinator Lab]
 +
* Overview of Weka & the Wekinator
 +
** [http://www.cs.waikato.ac.nz/ml/weka/ Weka home]
 +
** [http://code.google.com/p/wekinator/ Wekinator on Google code] and [http://wiki.cs.princeton.edu/index.php/ChucK/Wekinator/Instructions instructions]
 +
* A brief history of MIR
 +
** See also http://www.ismir.net/texts/Byrd02.html
 +
* Notes
 +
** CAL500 decoding
 +
for i in *.mp3; do echo $i; afconvert -d BEI16@44100 -f AIFF "$i"; done
 +
* Extract CAL 500 per-song features to .mat or .csv using features from today.  This will be used on lab for Friday. Copy it from the folder ccrma-gate.stanford.edu:/usr/ccrma/workshops/mir2011/cal500.tar (beware it's a 2Gb .tar file!) or grab the AIFF versions from ccrma-gate.stanford.edu:/usr/ccrma/workshops/mir2011/cal500_aiffs.tar (that's 16 GB)
  
 
== software, libraries, examples ==
 
== software, libraries, examples ==
Line 262: Line 224:
  
 
== Past CCRMA MIR Workshops and lectures==  
 
== Past CCRMA MIR Workshops and lectures==  
 +
* [http://ccrma.stanford.edu/wiki/MIR_workshop_2011 CCRMA MIR Summer Workshop 2011]
 +
* [http://ccrma.stanford.edu/wiki/MIR_workshop_2010 CCRMA MIR Summer Workshop 2010]
 
* [http://ccrma.stanford.edu/wiki/MIR_workshop_2009 CCRMA MIR Summer Workshop 2009]
 
* [http://ccrma.stanford.edu/wiki/MIR_workshop_2009 CCRMA MIR Summer Workshop 2009]
 
* [http://cm-wiki.stanford.edu/wiki/MIR_workshop_2008 CCRMA MIR Summer Workshop 2008]
 
* [http://cm-wiki.stanford.edu/wiki/MIR_workshop_2008 CCRMA MIR Summer Workshop 2008]
Line 279: Line 243:
  
 
Papers:
 
Papers:
 +
* ISMIR 2011 Proceedings: http://ismir2011.ismir.net/program.html
 
* Check out the references listed at the end of the Klapuri & Davy book
 
* Check out the references listed at the end of the Klapuri & Davy book
 
* Check out Papers listed on Pg 136-7 of  MIR Toolbox: http://www.jyu.fi/hum/laitokset/musiikki/en/research/coe/materials/mirtoolbox/userguide1.1
 
* Check out Papers listed on Pg 136-7 of  MIR Toolbox: http://www.jyu.fi/hum/laitokset/musiikki/en/research/coe/materials/mirtoolbox/userguide1.1

Latest revision as of 14:21, 26 June 2013

Intelligent Audio Systems: Foundations and Applications of Music Information Retrieval

Logistics

Workshop Title: Intelligent Audio Systems: Foundations and Applications of Music Information Retrieval

2013 MIR_workshop_2013

  • Participants:

Abstract

How would you "Google for audio", provide music recommendations based your MP3 files, or have a computer "listen" and understand what you are playing? This workshop will teach the underlying ideas, approaches, technologies, and practical design of intelligent audio systems using Music Information Retrieval (MIR) algorithms.

MIR is a highly-interdisciplinary field bridging the domains of digital audio signal processing, pattern recognition, software system design, and machine learning. Simply put, MIR algorithms allow a computer to "listen" and "understand or make sense of" audio data, such as MP3s in a personal music collection, live streaming audio, or gigabytes of sound effects, in an effort to reduce the semantic gap between high-level musical information and low-level audio data. In the same way that listeners can recognize the characteristics of sound and music - tempo, key, chord progressions, genre, or song structure - MIR algorithms are capable of recognizing and extracting this information, enabling systems to perform extensive sorting, searching, music recommendation, metadata generation, transcription, and even aiding/generating real-time performance.

This workshop is intended for: students, researchers, and industry audio engineers who are unfamiliar with the field of Music Information Retrieval (MIR). We will demonstrate the myriad of exciting technologies enabled by the fusion of basic signal processing techniques with machine learning and pattern recognition. Lectures will cover topics such as low-level feature extraction, generation of higher-level features such as chord estimations, audio similarity clustering, search, and retrieval techniques, and design and evaluation of machine classification systems. The presentations will be applied, multimedia-rich, overview of the building blocks of modern MIR systems. Our goal is to make the understanding and application of highly-interdisciplinary technologies and complex algorithms approachable.

Knowledge of basic digital audio principles is required. Familiarity with Matlab is desired. Students are highly encouraged to bring their own audio source material for course labs and demonstrations.

Workshop structure: The workshop will consist of half-day lectures, half-day supervised lab sessions, demonstrations, and discussions. Labs will allow students to design basic ground-up "intelligent audio systems", leveraging existing MIR toolboxes, programming environments, and applications. Labs will include creation and evaluation of basic instrument recognition, transcription, and real-time audio analysis systems.

Schedule: Lectures & Labs

Day 1: Introduction to MIR, Signal Analysis and Feature Extraction

Presenters: Jay LeBoeuf, Leigh Smith


Day 1: Part 1 Lecture 1 Part 1 Slides

  • Introductions
  • CCRMA Introduction - (Carr/Sasha). CCRMA Tour.
  • Introduction to MIR (What is MIR? Why are people interested? Commercial Applications of MIR)
  • Overview of a basic MIR system architecture
  • Timing and Segmentation: Frames, Onsets
  • Features: ZCR, Spectral moments; Scaling of feature data
  • Demo: Using simple heuristics and thresholds (i.e. "Why do we need machine learning?")
  • Classification: Instance-based classifiers (k-NN)
  • Information Retrieval Basics (Part 1)
    • Classifier evaluation (Cross-validation, training and test sets)


Day 1: Part 2 Lecture 2 Slides

  • Overview: Signal Analysis and Feature Extraction for MIR Applications (Historical: http://quod.lib.umich.edu/cgi/p/pod/dod-idx?c=icmc;idno=bbp2372.1999.356)
  • MIR Application Design
    • Audio input, analysis
    • Statistical/perceptual processing
    • Data storage
    • Post-processing
  • Windowed Feature Extraction
    • I/O and analysis loops
  • Feature-vector design (Overview: http://www.create.ucsb.edu/~stp/PostScript/PopeHolmKouznetsov_icmc2.pdf)
    • Kinds/Domains of Features
    • Application Requirements (labeling, segmentation, etc.)
  • Time-domain features (MPEG-7 Audio book ref)
    • RMS, Peak, LP/HP RMS, Dynamic range, ZCR
  • Frequency-domain features
    • Spectrum, Spectral bins
    • Spectral measures (statistical moments)
    • Pitch-estimation and tracking
    • MFCCs
  • Spatial-domain features
    • M/S Encoding, Surround-sound Processing Frequency-dependent spatial separation, LCR sources
  • Other Feature domains
    • Wavelets, LPC


  • Application: Instrument recognition and drum transcription / Using simple heuristics and thresholds (i.e. "Why do we need machine learning?")


Lab 1:

Students who need a personal tutorial of Matlab or audio signal processing will split off and received small group assistance to bring them up to speed.

  • REMINDER: Save all your work, because you may want to build on it in subsequent labs.

Day 2: Beat-finding and Rhythm Analysis

Presenter: Leigh Smith

Lecture 3 Slides A list of beat tracking references cited

  • Onset-detection: Many Techniques
    • Time-domain differences
    • Spectral-domain differences
    • Perceptual data-warping
    • Adaptive onset detection
  • Beat-finding and Tempo Derivation
    • IOIs and Beat Regularity, Rubato
      • Tatum, Tactus and Meter levels
      • Tempo estimation
    • Onset-detection vs Beat-detection
      • The Onset Detection Function
    • Approaches to beat tracking & Meter estimation
      • Autocorrelation
      • Beat Spectrum measures
      • Multi-resolution (Wavelet)
    • Beat Histograms
    • Fluctuation Patterns
    • Joint estimation of downbeat and chord change


Lab 2:

Day 3: Music Information Retrieval in Polyphonic Mixtures

Presenter: Steve Tjoa

Lecture/Lab 3

  • Music Transcription and Source Separation
  • Nonnegative Matrix Factorization
  • Sparse Coding
  • Locality Sensitive Hashing

Day 4: Pitch, Chroma, More Classification

Presenters: Steve Tjoa, Oscar Celma (Gracenote)

Lecture 4

  • Features:
    • Monophonic Pitch Detection
    • Polyphonic Pitch Detection
    • Pitch representations (Tuning Histograms, Pitch and Pitch Class Profiles, Chroma)
  • Analysis:
    • Dynamic Time Warping
    • Hidden Markov Models
    • Harmonic Analysis/Chord and Key Detection
  • Applications
    • Audio-Score Alignment
    • Cover Song Detection
    • Query-by-humming
    • Music Transcription
  • Music Recomendation
    • Overview of music recommendation. What's hard about it.
    • Some statistics and observations about the music industry and the need for recomendation.
    • Point-Counterpoint: Should we bother with content-based analysis.

Lab 4

Day 5: Michael Mandel, Jay LeBoeuf, Steve Tjoa, Leigh Smith

Guest Lecture: Autotagging, Michael Mandel

References:

  • Time permitting:
    • Advanced features
    • Sparse coding
    • Using musical structure.

Lab 5

Below are some links to functions and matlab code for key estimation, chord recognition, and GMMs:



Bonus Lab material

for i in *.mp3; do echo $i; afconvert -d BEI16@44100 -f AIFF "$i"; done

  • Extract CAL 500 per-song features to .mat or .csv using features from today. This will be used on lab for Friday. Copy it from the folder ccrma-gate.stanford.edu:/usr/ccrma/workshops/mir2011/cal500.tar (beware it's a 2Gb .tar file!) or grab the AIFF versions from ccrma-gate.stanford.edu:/usr/ccrma/workshops/mir2011/cal500_aiffs.tar (that's 16 GB)

software, libraries, examples

Applications & Environments

Machine Learning Libraries & Toolboxes

Optional Toolboxes

Supplemental papers and information for the lectures...

Past CCRMA MIR Workshops and lectures

References for additional info

Recommended books:

  • Data Mining: Practical Machine Learning Tools and Techniques, Second Edition by Ian H. Witten , Eibe Frank (includes software)
  • Netlab by Ian T. Nabney (includes software)
  • Signal Processing Methods for Music Transcription, Klapuri, A. and Davy, M. (Editors)
  • Computational Auditory Scene Analysis: Principles, Algorithms, and Applications, DeLiang Wang (Editor), Guy J. Brown (Editor)
  • Speech and Audio Signal Processing:Processing and perception of speech and music Ben Gold & Nelson Morgan, Wiley 2000

Prerequisite / background material:

Papers:

Other books:

  • Pattern Recognition and Machine Learning (Information Science and Statistics) by Christopher M. Bishop
  • Neural Networks for Pattern Recognition, Christopher M. Bishop, Oxford University Press, 1995.
  • Pattern Classification, 2nd edition, R Duda, P Hart and D Stork, Wiley Interscience, 2001.
  • "Artificial Intelligence: A Modern Approach" Second Edition, Russell R & Norvig P, Prentice Hall, 2003.
  • Machine Learning, Tom Mitchell, McGraw Hill, 1997.

Interesting Links:

Audio Source Material

OLPC Sound Sample Archive (8.5 GB) [1]

http://www.tsi.telecom-paristech.fr/aao/en/category/database/

RWC Music Database (n DVDs) [available in Stanford Music library]

RWC - Sound Instruments Table of Contents

http://staff.aist.go.jp/m.goto/RWC-MDB/rwc-mdb-i.html

Univ or Iowa Music Instrument Samples

https://ccrma.stanford.edu/wiki/MIR_workshop_2008_notes#Research_Databases_.2F_Collections_of_Ground_truth_data_and_copyright-cleared_music

MATLAB Utility Scripts

http://ccrma.stanford.edu/~kglee/kaist_summer2008_special_lecture/