[MEI-L] PhD position on Music Information Retrieval / Music transcription. CNAM, Paris

philippe philippe.rigaux at cnam.fr
Wed Jul 5 19:34:17 CEST 2017

(apologies for cross-posting)


In the context of a collaborative research program between the Vertigo/Database 
team at the laboratory Cedric at Conservatoire des Arts et MĂ©tiers (CNAM) and 
Mc Gill and Montreal Universities,  we offer a PhD position dedicated to the development 
of prior models of languages of music scores for automated music transcription and 
music information retrieval.

Supervisors of the position: Philippe Rigaux and Florent Jacquemard
Contacts for applications: philippe.rigaux at cnam.fr, florent.jacquemard at inria.fr

The PhD work will aim at studying appropriate prior models for 
representing languages of admissible music scores and guiding the generation of 
relevant notation in automated music transcription. For instance, such a target 
language model can be combined with a performance model in a transcription
procedure that returns one or more scores corresponding to a given interpretation.

This doctoral work will require methodologies coming from formal language 
theory and natural languages processing in order to design rich language models able 
to take into account specific elements of notation (grouping of notes, ties, dots, 
grace notes…) and underlying metric constraints (for example invariants on the sum of 
note durations within a beat or a measure). 
A particular attention will be dedicaced to the application of models and transcription 
methods to music information retrieval issues such document search, indexing and 
classification in music digital libraries such as Neuma http://neuma.huma-num.fr.

The PhD position is part of Munir, a coordinated research program devoted to the search 
and analysis of music that has been encoded into representative symbolic music format 
(scores), and to the development of collections and tools to demonstrate the practical 
impact and to disseminate the project's results. 

This PhD grant is a 3-year fellowship with an annual evaluation. The PhD student will 
work at the Cedric laboratory of the CNAM institute https://cedric.cnam.fr
The applicants should hold a Master degree in computer science which would formally 
entitle him/her to embark on a doctorate for the academic course 2017-2018 at this institute.
Knowledge in music representations (audio or symbolic) would be much appreciated.

More information about the mei-l mailing list