MIR workshop 2015

From CCRMA Wiki
Revision as of 13:49, 10 July 2015 by Kiemyang (Talk | contribs) (software, libraries, examples)

Jump to: navigation, search

Intelligent Audio Systems: Foundations and Applications of Music Information Retrieval

Logistics

Abstract

How would you "Google for audio", provide music recommendations based on your MP3 files, or have a computer "listen" and understand what you are playing?

This workshop will teach such 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 to, understand, and 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 sort, search, recommend, tag, and transcribe music, possibly in real time.

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 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 Python is desired but not required. 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 audio analysis systems.

Schedule

Instructional material can be found at musicinformationretrieval.com (read only) or on GitHub (full source).

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

Lecture

Introductions

  • CCRMA Introduction - (Nette, Fernando).
  • Introduction to MIR (What is MIR? Why MIR? Commercial applications)
  • Basic MIR system architecture
  • Timing and Segmentation: Frames, Onsets
  • Classification: Instance-based classifiers (k-NN)

Overview: Signal Analysis and Feature Extraction for MIR Applications

MFCCs sonified

  • Original track ("Chewing Gum") [1]
  • MFCCs only [2]


Lab

Understanding Audio Features Through Sonification

Day 2: Beat, Rhythm, Pitch and Chroma Analysis

List of beat tracking references

Onset Detection

  • Time-domain differences
  • Spectral-domain differences
  • Perceptual data-warping
  • Adaptive onset detection

Beat and Tempo

  • IOIs and Beat Regularity, Rubato
  • Tatum, Tactus and Meter levels
  • Tempo estimation
  • Onset-detection vs Beat-detection
  • The Onset Detection Function
  • Beat Histograms
  • Fluctuation Patterns
  • Joint estimation of downbeat and chord change

Approaches to Beat Tracking and Meter Estimation

  • Autocorrelation
  • Beat Spectrum measures
  • Multi-resolution (Wavelet)

Pitch and Chroma

  • 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

Lab

Part 1: Tempo Extraction

Part 2: Add in MFCCs to classification and test w Cross validation

Bonus Slides: Temporal & Harmony Analysis

Day 3: Machine Learning, Clustering and Classification

Lecture

Classification: Unsupervised vs. Supervised, k-means, GMM, SVM

Lab

MFCC, K-Means Clustering

Day 4: Music Information Retrieval in Polyphonic Mixtures

Lecture

Music Transcription and Source Separation

  • Nonnegative Matrix Factorization
  • Sparse Coding

Evaluation Metrics for Information Retrieval

Lab

Lab 4 Description

References:

Day 5: Deep Belief Networks and Wavelets

Lecture 10: Steve Tjoa, Introduction to Deep Learning Slides

Lecture 11: Leigh Smith, An Introduction to Wavelets Slides

[ https://ccrma.stanford.edu/workshops/mir2014/fann_en.pdf Neural Networks made easy]

Lunch at The Oasis

Klapuri eBook: http://link.springer.com/book/10.1007%2F0-387-32845-9

software, libraries, examples

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) [3]

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/

MIR_workshop_2014

Bonus Lab Material from Previous Years (Matlab)

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)