Computational models for the analysis, characterization and synthesis of flamenco music

In our research, we investigate how music information retrieval techniques can support the analysis, characterization and generation of flamenco music.

Over the last few years, we have worked on MIR topics applied to flamenco: automatic transcription (Kroher and Gómez, 2016) (Gómez and Bonada, 2013), melodic similarity (Kroher et al., 2014), singer identification (Kroher and Gómez, 2014), melodic pattern detection (Pikrakis et al., 2014), singing voice synthesis and web mining (Oramas et al., 2015). In addition, we have published a set of datasets, some of them available at the MTG web site and others at the COFLA website (see below). 

The COFLA (Computational Analysis of Flamenco Music) project (Feb 2014-Feb 2018. COFLA2: Análisis Computacional de la Música Flamenca), Proyectos de Excelencia de la Junta de Andalucía (P12-TIC-1362) provides an interdisciplinary forum to work on flamenco music, from the discipines of anthropology, mathematics, musicology and engineering. COFLA is funded by the Andalusian Government. 

More info can be found on the COFLA project website.