Artificial intelligence to determine what music on TV is subject to copyright


Music from television productions is a major source of revenue for the music industry through copyright. Copyright can only be exercised when the music is audible enough, but this is difficult to determine depending on the context. For example, our perception may change when other sounds, such as conversation, overlap.

In order to provide a tool to solve this problem, the UPF spin-off BMAT and the Music Technology Group (MTG) are collaborating to develop a new technology that automatically and objectively establishes the degree of audibility of the background music and determines so when copyright should be enforced.

“Our goal is to provide the international music industry with an innovative service that helps distribute copyright among its legitimate owners in a more equitable, efficient and transparent manner.”

The project, called LoudSense, received an Innotec grant in the 2020 announcement. This program, which promotes technology transfer, funds R&D projects carried out jointly by companies and research groups labeled TECNIO.

The grant, awarded by ACCIÓ, the Catalan government’s agency for business competitiveness, granted BMAT and MTG 146,197 euros to develop their technology over two years.

“We aim to provide the international music industry with an innovative service that allows copyright to be distributed among its rights holders in a more equitable, efficient and transparent manner”, summarizes Gonçal Calvo, in charge of innovation at BMAT.

Technology development

This project will consist of two phases. In the first, the MTG will perform a series of experiments related to human perception of sound to establish what characteristics music must have to be considered audible depending on the context. The results of this study will help researchers establish criteria and generate a dataset.

In a second phase, BMAT will take advantage of all the information generated during the experiments to develop an algorithm based on artificial intelligence that will automatically determine when the music is audible, which will make it possible to objectively establish whether to generate rights to author.

LoudSense will “have a very big impact on the market because it will solve a problem in the music industry which is very controversial and for which no one is currently able to offer a service like this”

LoudSense will “have a very significant impact on the market because it will solve a problem in the music industry which is very controversial and for which currently no one is able to offer a service like this”, assures Calvo.

An industrial doctorate success story

The seed of LoudSense was born thanks to the industrial doctorate program of the Catalan government, which encourages the realization of doctoral theses within the framework of collaborative applied research projects between the company and the universities. In the 2016 edition, BMAT and MTG cooperated in this program with a thesis on a technology capable of automatically detecting and categorizing background and foreground music.

“The industrial doctorate was successful because the university-industry collaboration worked very well and laid the foundation for the launch of LoudSense. In addition, the investigator, who was then a doctoral student, is now a doctor and continues to work on the project, ”explains Sonia Espí, promoter of the MTG.

Benefits of collaboration

This knowledge transfer project has several advantages from the point of view of the MTG: “First, through the identification of problems and market needs, we are able to maximize the impact of our research in the real world, by the music industry occurrence; secondly, working with companies will help us improve our technologies and our capabilities, ”Judge Espí.

The head of innovation of BMAT adds that the collaboration with the research group will allow the company to “take up a challenge which goes beyond the current state of the art” and to improve its services and technologies. music monitoring “to gain a competitive advantage and market share”. .

With the support of ACCIÓ


Warning: AAAS and EurekAlert! are not responsible for the accuracy of any press releases posted on EurekAlert! by contributing institutions or for the use of any information via the EurekAlert system.


About Author

Comments are closed.