Home About Deliverables Publications FB twitter

Dig That Lick

Analysing large-scale data for melodic patterns in jazz performances

Project News:


Save this date: 24 April 2020 for the final workshop of the Dig that Lick project, to be held in Paris. More information to follow shortly.


An overview of the Dig that Lick project was presented at the Round 4 Digging Into Data Challenge Conference at the National Science Foundation in Alexandria, Virginia, USA. The slides are available here.


New Release: Jazz Pattern Search Website

We are happy to announce the new release of the Dig That Lick Pattern Similarity Search website. It is one of the major deliverables of our 2-year research project. It can be found here: https://dig-that-lick.hfm-weimar.de/similarity_search/

Key features:

  • Currently we support four melodic databases:
    1. The new DTL1000 database, comprising 300000 tone events in 1736 monophonic solos from over 600 jazz tunes spanning the 100 years of jazz history. The solos have been extracted automatically from audio using a newly developed CRNN-based algorithm specialised for jazz.
    2. The well-known Weimar Jazz Database with about 200000 tone events from 456 monophonic solos by 78 jazz masters.
    3. The Charlie Parker Omnibook with about 18000 tones taken from 52 solos by the co-inventor of bebop.
    4. The Essen Folk Song Collection, comprising about 350000 notes from 7352 folk songs.
  • Similarity search can be carried out using interval, refined contour, and pitch patterns (n-grams). The underlying similarity measure is based on the Levenshtein Distance, which gives a reasonable approximation to true perceptual similarity. Various user-definable search parameters, a virtual keyboard for query input, and extensive metadata filters are also available.
  • The result list shows all pattern instances for a given query in the user-defined similarity range with essential metadata and audio snippets for quick aural control. The search results can be grouped by performer (or folk song collection) and by pattern. Extra information for each pattern instance can be displayed by the user according to his or her needs. Result sets can be exported to CSV files by a single click.
  • Furthermore, we provide several visualisation options for result sets such as a pattern timeline and various kinds of pattern networks.
  • Global and personal search histories are available for quick retrieval of previous searches and for exploration of other users’ queries.
  • And, of course, there is extensive documentation available.

More information is available in our Late-Breaking Demo paper at ISMIR 2019: Dig that Lick: Exploring Melodic Patterns in Jazz Improvisation by Frank Höger, Klaus Frieler, Martin Pfleiderer and Simon Dixon.


Dig That Lick at Rhythm Changes

We presented a panel on our project at Rhythm Changes - the main European academic conference on jazz. The panel included following talks:

  • A Technical Primer for Big Data Jazz Studies
  • Towards a history of melodic patterns in jazz performance
  • What We Are Digging Out of the Data
  • Chasing the Trane: Quantifying the Social Journey of a Coltrane Solo

New release: Dig That Pattern Similarity Search in the Weimar Jazz Dataset

We are delighted to announce the release of our next cool web tool, the Dig That Pattern Similarity Search. It allows to search the Weimar Jazz Database not only for exact interval and pitch patterns but also for similar ones, which opens completely new possibilites. It includes very cool visualizations, such as pattern networks and timelines, and many more features.

Get started with a true bebop-ish example: -1,-2,-1,3,3,3,-1,-2


Search for and explore jazz solo patterns in the Weimar Jazz Dataset

We proudly present two brand new web applications dealing with patterns in jazz solos. The first is an interactive app called the Pattern History Explorer. There, you can explore 653 common patterns with 11,630 instances in the Weimar Jazz Database from many different angles including score and audio snippets. The second is a web interface for pattern search in the Weimar Jazz Database with many cool features. If you want to know more about all this, you can read our accompanying paper Two web applications for exploring melodic patterns in jazz solos , which got accepted for the ISMIR 2018 conference in Paris.


Two papers from our project were accepted for the annual conference of the International Scociety for Music Information Retrieval (ISMIR2018) - the leading worldwide conference in this research area:

"Main Melody Extraction with source-filter NMF and C-RNN"

by D. Basaran, S. Essid and G. Peeters as well as

"Two web applications for exploring melodic patterns in jazz solos"

by Klaus Frieler, Frank Höger, Martin Pfleiderer, and Simon Dixon

Please see our Publications page for more details.

3-4/05/2018 Dig That Lick project meeting at IRCAM, Paris, France

Deliverable D1.1: Licks in the Literature of Jazz Research

by Gabriel Solis

Having completed a review of academic jazz literature from the 1970s to the present we have found considerable range of conception and use of the general topic of patterns in improvisation and a number of potentially testable hypotheses that our data set and analytical work might address. This paper offers a synthesis of key issues from that research, including: definitions and terminology to describe patterns; disciplinary frameworks in which the question of pattern is engaged; primary areas of debate; a note on the volume of datasets needed for this research; and testable hypotheses to consider.

7/12/2017 "Inside the Jazzomat. New perspectives for jazz research" by Martin Pfleiderer et al. has been published and is available online. This is the publication resulting from the Jazzomat project which was in many aspects the predecessor of Dig That Lick.

Welcome to the web site of the project: Dig That Lick: Analysing large-scale data for melodic patterns in jazz performances, a collaboration between six different universities across four countries, funded by the Trans-Atlantic Program Digging into Data Challenge (see announcement here).

The project started on 1 Oct 2017 and is funded until 30 Sep 2019.

Read more about the project on our About page.