Laurent Pugin

Goal-directed evaluation for the improvement of optical music recognition on early music prints

Optical music recognition (OMR) systems are promising tools for the creation of searchable digital music libraries. Using an adaptive OMR system for early music prints based on hidden Markov models, we leverage an edit distance evaluation metric to improve recognition accuracy. Baseline results are computed with new labeled training and test sets drawn from a diverse group of prints. We present two experiments based on this evaluation technique. The first resulted in a significant improvement to the feature extraction function for these images. The second is a goal-directed comparison of several popular adaptive binarization algorithms, which are often evaluated only subjectively. Accuracy increased by as much as 55% for some pages, and the experiments suggest several avenues for further research.

Reducing costs for digitising early music with dynamic adaptation

Optical music recognition (OMR) enables librarians to digitise early music sources on a large scale. The cost of expert human labour to correct automatic recognition errors dominates the cost of such projects. To reduce the number of recognition errors in the OMR process, we present an innovative approach to adapt the system dynamically, taking advantage of the human editing work that is part of any digitisation project. The corrected data are used to perform MAP adaptation, a machine-learning technique used previously in speech recognition and optical character recognition (OCR). Our experiments show that this technique can reduce editing costs by more than half.

Digital preservation and access strategies for musical heritage: the Swiss RISM experience

Purpose – The purpose of this paper is to present a new web‐based cataloguing system for the global music bibliography project, Repertoire International des Sources Musicales (RISM), and discuss the implications for the manipulation and discovery of musical heritage materials.Design/methodology/approach – The paper is designed to illustrate the workflow and tools used in creating a global musical catalogue, and to present the experiences of the Swiss RISM working group in developing new tools and re‐thinking traditional music bibliography tools.Findings – The new tools developed present a further decrease in latency between source cataloguing and availability to users by integrating both the cataloguing and exploration interfaces into a single web application.Research limitations/implications – For music researchers, the opportunity to search and manipulate a global musical source database opens up new possibilities for data‐driven computational musicology and analysis.Originality/value – This paper repor...

Automatic Score Extraction with Optical Music Recognition (OMR)

Optical music recognition (OMR ) describes the process of automatically transcribing music notation from a digital image. Although similar to optical character recognition (OCR ), the process and procedures of OMR diverge due to the fundamental differences between text and music notation, such as the two-dimensional nature of the notation system and the overlay of music symbols on top of staff lines. The OMR process can be described as a sequence of steps, with techniques adapted from disciplines including image processing, machine learning, grammars, and notation encoding. The sequence and specific techniques used can differ depending on the condition of the image, the type of notation, and the desired output.

Instrument distribution and music notation search for enhancing bibliographic music score retrieval

Because of the unique characteristics of music scores, searching bibliographical music collections using traditional library systems can be a challenge. In this paper, we present two specific search functionalities added to the Swiss RISM data-base and describe how they improve the user experience. The first is a search functionality for instrument and vocal part distribution that leverages coded information available in the MarcXML records of the database. It enables scores for precise ensemble distribution to be retrieved. The second is a search functionality of music notation excerpts transcribed from the beginning of the pieces, known as music incipits. The incipit search is achieved using a well-known music information retrieval (MIR) tool, Themefinder. A novelty of our implementation is that it can operate at three different levels (pitch, duration and metric), singularly or combined, and that it is performed through a specifically-developed intuitive graphical interface for note input and parameter selection. The two additions illustrate why it is important to take into consideration the particularities of music scores when designing a search system and how MIR tools can be beneficially integrated into existing heterogeneous bibliographic score collections.