MUSIC INFORMATION RETRIEVAL: AN OVERVIEW Jordan B. L. Smith - - PowerPoint PPT Presentation

music information retrieval an overview
SMART_READER_LITE
LIVE PREVIEW

MUSIC INFORMATION RETRIEVAL: AN OVERVIEW Jordan B. L. Smith - - PowerPoint PPT Presentation

MUSIC INFORMATION RETRIEVAL: AN OVERVIEW Jordan B. L. Smith Mathemusical Conversations Study Day 12 February 2015 Raffles Institution OUTLINE A bit about me Applications (the fun stuff!) How do they do it? National Institute of


slide-1
SLIDE 1

MUSIC INFORMATION RETRIEVAL: AN OVERVIEW

Jordan B. L. Smith Mathemusical Conversations Study Day 12 February 2015 Raffles Institution

slide-2
SLIDE 2

OUTLINE

  • A bit about me
  • Applications (the fun stuff!)
  • How do they do it?
slide-3
SLIDE 3
  • utline
  • A bit about me
  • Applications (the fun stuff!)
  • How do they do it?

Montreal Boston Harvard Music and Physics Music Technology McGill Los Angeles Operations Research Engineering USC London Queen Mary University of London Electrical Engineering and Computer Science Tsukuba National Institute of Advanced Industrial Science and Technology Media Interaction Group .

slide-4
SLIDE 4

MIR APPLICATIONS

Music Fingerprinting Query By Humming Music Similarity & Recommendation Playlist Generation

slide-5
SLIDE 5

MIR APPLICATIONS

Microsoft Songsmith commercial: https://www.youtube.com/watch?v=3oGFogwcx-E

slide-6
SLIDE 6

MIR APPLICATIONS

Microsoft Songsmith commercial: https://www.youtube.com/watch?v=3oGFogwcx-E

slide-7
SLIDE 7

MIR APPLICATIONS

  • Recognition (e.g., Shazam, SoundHound)
  • Recommender systems (e.g., Spotify, Pandora,

YouTube, iTunes Genius)

  • Play (e.g., Microsoft Songsmith)
  • Metadata management (e.g., Gracenote, MusicBrainz)
slide-8
SLIDE 8

echonest.com

slide-9
SLIDE 9

http://developer.echonest.com/docs/v4/track.html

slide-10
SLIDE 10

http://developer.echonest.com/docs/v4/_static/AnalyzeDocumentation.pdf

slide-11
SLIDE 11

AUDIO FEATURE EXTRACTION

  • To do:
  • demonstrate audio spectrum analysis using Sonic

Visualiser

  • show log scaling
  • mention Pythagoras, who founded music theory

and science with a single stroke

slide-12
SLIDE 12

STRUCTURAL ANALYSIS

  • Goal: find large-scale sections from audio
slide-13
SLIDE 13

STRUCTURAL ANALYSIS

Masataka Goto, SmartMusicKIOSK: Music-playback interface based on chorus-section detection method, The Journal of the Acoustical Society of America, Vol.115, No.5, Pt.2, p.2494, May 2004. (Invited Paper of The 147th Meeting of the Acoustical Society of America)

slide-14
SLIDE 14

MINI-APPLICATION: STRUCTURE

Davies, M.E.P ., Hamel, P . ; Yoshii, K. ; Goto, M. 2014. AutoMashUpper: Automatic Creation of Multi-Song Music Mashups. Transactions on Audio Speech and Language Processing 22 (12): 1726–1737.

slide-15
SLIDE 15

ECHO NEST REMIX

  • Gives full rhythmic and structural analysis
  • Enables automatic remixing
  • Example: Girl Talk in a box
slide-16
SLIDE 16

SUMMARY

  • Music Information Retrieval includes:
  • retrieving musical documents from queries
  • interpreting social data about music
  • extracting information from music
  • …and you can do it!
slide-17
SLIDE 17

THANK YOU!