Conference Agenda

The semantics of notation systems can naturally be meta-modelled

as a network of transformations, starting with the syntactic elements

of the notation and ending with the parameters of an execution. In

this context, a digital encoding format for music notation can be seen

as selecting a subset of the data nodes of this network for storage,

leaving others to evaluation. For such a selection, semantic properties

are defined which have impact on the practical costs of maintenance,

migration, extension, etc.

This paper examines methods for encoding musical scores used for the Hichiriki instrument in gagaku, traditional Japanese court music. Gagaku is Japan's oldest musical tradition but its scores use ambiguous notation, posing challenges for computational analysis. The proposed encoding method centers on shōga (sung mnemonic lyrics) which appear most frequently. The elements of tetsuke (fingering) and hyōshi (rhythmic cycles) accompany shōga in markup. Efficient workflow utilizes Python for initial markup and a Stream Deck device for rapid tagging while reviewing images of original scores. Encoded data created for 93 short and medium pieces provides a structured foundation to computationally study gagaku. Future work will expand encoding to parts for other wind instruments like shō and ryūteki.

This paper explores Wikidata’s transformative potential as a language-agnostic tool for categorizing musical instruments. Traditional taxonomy systems like Hornbostel-Sachs (HBS) or the Library of Congress (LOC) face challenges due to language-specific constraints and a lack of provisions for local terminology. Wikidata, a collaborative knowledge base, serves as a solution, allowing structured data storage and linking across diverse topics. It employs a hierarchical structure with unique identifiers (Q-ids and P-ids) for items and properties, organizing them within a conceptual framework.

Despite challenges arising from Wikidata’s open-source nature, its machine-readable format and multilingual support provide a robust solution. It addresses inconsistencies, such as incomplete musical instrument holdings, by offering a more inclusive musical knowledge ecosystem. Users can explore, contribute, and query information in their preferred languages, fostering a diverse landscape.

Wikidata also acts as a tool for archiving existing databases, notably used by museums like the Metropolitan Museum of Art and the British Museum. Overcoming linguistic isolation, it links unique items to identifiers, facilitating queries for historical instruments. This is especially crucial for identifying linguistically isolated historical instruments documented in encyclopedias or articles.

Contrasting with traditional hierarchical systems, Wikidata’s item-specific approach offers greater flexibility and access. The paper explores Wikidata’s collaborative, multilingual, and machine-readable framework, highlighting its role in projects like LinkedMusic’s Virtual Instrument Museum (VIM), which utilizes Wikidata to create a comprehensive virtual database of musical instrument names, expanding hierarchies and terminology. The synergy between Wikidata and VIM enriches musical knowledge dissemination, engaging specialists and nonspecialists alike.

In conclusion, the paper underscores Wikidata’s potential impact on musical instrument encoding, promoting linguistic diversity through machine-readable, language-agnostic identifiers. Collaborative projects, exemplified by VIM, demonstrate Wikidata's ability to bridge language gaps, contributing to a globally enriched understanding of diverse musical instruments.

Music encoding is a extensive field of research and therefore it needs a unification of encoding rules according to the music it deals with.

The proposal advocates the use of Optical Musica Reconigtion for the automatic encoding of jazz lead sheets. For them it will be necessary to explore the musicological complexities of jazz, identifying challenges associated with the improvisational nature, handwritten notations and notational ambiguities, harmonic complexities. To adapt these musical and notational challenges to OMR technology based on the creation of a diverse dataset composed of handwritten sources to enrich the information to train Artificial Intelligence (AI)-based Music Information Retrieval models.

The main goal is to achieve an efficient automatic encoding, contributing to the preservation and dissemination of jazz musical heritage.