The Audio Commons Initiative aims to foster creative use of audio content, through favouring interoperability between audio tools and content repositories. For this reason we designed a common data model for audio content, the Audio Commons Ontology. This model has been published and documented and will soon be integrated in multiple tools of the Audio Commons Ecosystem, such as the Audio Commons Mediator, the tool that integrates some existing content repositories (Jamendo, Freesound, and Europeana) offering a common web API.
Figure 1. Diagram of the Audio Commons Ontology
The Audio Commons Ontology is inspired by (and related to) the Music Ontology and the Functional Requirements for Bibliographic Records (FRBR) model. It offers a way to represents the relationships and metadata that are useful while searching for audio content in multiple repositories. It defines concepts as AudioClips, AudioCollections, AudioCategories, and AudioFiles. It further defines their relationships and the metadata that may be used to identify, describe, and classify them. It is formally described using the Web Ontology Language (OWL) and documented for use in its web page (https://w3id.org/ac-ontology/aco).
The data model described by the Audio Commons Ontology is central to the Audio Commons Ecosystem and it is expected to be adopted by the services, tools, and agents that compose the system. The Audio Commons consortium is working on enhancing the existing tools accordingly and develop new tools that will be using the ontology. One important example is the Audio Commons Mediator, a piece of software that integrates three existing content repositories (Jamendo, Freesound, and Europeana), offering a common web API to access them in order to look for audio content in any of the three. In the context of the mediator, the ontology will be used for the dual role of being:
- the common data model used for integration;
- a way to express in the results rich data relationships that cannot be represented in the current version of the mediator.
Furthermore, the Audio Commons Ontology aims to represent audio content and its categories and relationships beyond the Audio Commons Initiative. A scientific paper describing the ontology will be presented in October at the International Semantic Web Conference (ISWC) in Monterey, CA, USA.