CANTOSTREAM

Computational ANalysis of TOnal STRuctures in EArly Music

In the 16th and 17th centuries, there was a significant shift from modality, which focuses on melodic structures, to harmonic tonality, which emphasizes chord progressions. Musicologists find it difficult to understand this transition by close-reading individual compositions. Identifying long-term historical patterns can be challenging with traditional musicological methods.

The CANTOSTREAM project adopts a big data approach. It aims to develop machine learning methods for distant listening to a large collection of music. This approach seeks to provide deeper insights into the historical development of tonal structures during these centuries.

In Boundaries we study the segmentation of renaissance music by humans and segmentation algorithms. In Fuzzy frequencies we invesitgate the possibilities of pitch extraction from audio for studying tonal structures. In Corpora we present the dissiculties in creating a representative corpus. In Team we present the project team.