Back TELMI facilitates music education through technology and artificial intelligence

TELMI facilitates music education through technology and artificial intelligence

Coordinated by Rafael Ramírez, director of the Music and Machine Learning Lab of the Music Technology Group, this  European Commission Research and Innovation action has just been completed after a three-year journey.

01.03.2019

 

In the traditional teacher-student model, learning to play a musical instrument often entails brief contact with the teacher and long periods of self-study by the learner which requires great perseverance by the aspiring musician leading to high dropout rates. The results of the TELMI project have provided students of music with technology to facilitate their learning.

TELMI (Technology-Enhanced Learning of Musical Instrument Performance) is a Research and Innovation action that has just been completed and was coordinated by Rafael Ramírez, head of the Music and Machine Learning Lab of the Music Technology Group (MTG) at the Department of Information and Communication Technologies (DTIC) of UPF, thanks to funding by the European Commission.  It started in February 2016 and was carried out over 36 months. It involved a consortium of academic partners and professionals in the field of music: Pompeu Fabra University (coord.), University of Genoa (Italy), Royal College of Music in London (UK) together with technology companies: HIGHSKILLZ and SAICO INTELLIGENCE, S.L.

A new way to study and learn a musical instrument

Taking the violin as a case study, the main goal of the TELMI project was to study how we learn to play musical instruments, from a pedagogical and scientific perspective, creating new helpful, interactive, self-learning content, i.e., systems to complement traditional teaching.  

The project has designed and implemented new interaction paradigms for music learning and training based on multimodal techniques (audio, image, video and movement); from a pedagogical point of view, it has evaluated the effectiveness of these new paradigms based on the results obtained and has developed new multimodal interactive music learning prototypes for students and teachers. It has also generated a reference database that is available to the public with recordings for online learning and to promote social interaction among students.

Artificial intelligence techniques have been applied to detect patterns and best practices in music education

For example, with the help of technology, 3D cameras were used to detect position and movement in learning the violin that tell the student, in the absence of the teacher, if their position and gestures with the instrument and bow are correct; and also while studying the technique of a particular performer or conductor. Artificial intelligence techniques have also been applied to detect patterns and best practices in music education.

“With violin learning as a case study, from a scientific and pedagogical perspective, a new paradigm has been achieved that combines interactive self-learning and augmented feedback, and provides systems to complement traditional teaching”, explained Rafael Ramírez, coordinator of this international project.

The project results will act as the basis for the development of music learning systems that improve interaction between students and teachers, students’ practice and, moreover, offer the possibility of bringing music education to a wider audience.

Multimedia

SDG - Sustainable Development Goals:

Els ODS a la UPF

Contact