A Computer System to Control Affective Content in Music Production



Music is a ubiquitous media in our lives, used in many contexts.
The possibility to select appropriate affective music can be helpful
to adapt music to our emotional interest. Our work intends to design a
system to control affective content in music production. This is done by
taking into account a knowledge base with mappings between affective
states (happiness, sadness, etc.) and music features (rhythm, melody,
etc.). The knowledge base is grounded on background knowledge from
Music Psychology. Our system starts with the reception of an emotional
description specified by the user. Next, mappings are selected from the
knowledge base, according to the emotional description. Music is retrieved
from a music base (recorded sound and MIDI files) according to
similarity metrics between music features (of mappings and music base).
Afterward, selected music can be subject to transforming, sequencing
and remixing algorithms and then played. The inclusion of third party
composition software is also envisaged. To assess the system, listener
emotional state can be analysed using psychophysiological or self-report


Music Computing


Music Computing


Portuguese Conference on Artificial Intelligence, December 2007

PDF File

Cited by

No citations found