Bibliographic citations
García, E., (2020). Desarrollo de un sistema compositor basado en sistemas expertos y redes neuronales para la generación de acompañamiento musical [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/21641
García, E., Desarrollo de un sistema compositor basado en sistemas expertos y redes neuronales para la generación de acompañamiento musical [Tesis]. PE: Universidad Nacional de Ingeniería; 2020. http://hdl.handle.net/20.500.14076/21641
@misc{renati/711279,
title = "Desarrollo de un sistema compositor basado en sistemas expertos y redes neuronales para la generación de acompañamiento musical",
author = "García Miranda, Erick Franz",
publisher = "Universidad Nacional de Ingeniería",
year = "2020"
}
In this research work, composer system is proposed that generates a musical accompaniment from an existing melody. The accompaniment consists of a harmony and a bass line, generated by an expert system and a recurrent neural network of long-term memory (LSTM) respectively. The expert system was built based on the knowledge of various musical experts in terms of harmonization and the neural network was trained with alternative rock and funk songs, using resistant propagation (Rprop). Also, the main architecture of the system was developed in Jython Music, while the inference engine of the expert system was developed in Java and the neural network was trained in Python. In addition, an opinion poll was conducted that evaluates songs generated by the composer system and songs from advertising campaigns in terms of how good and original they are. This survey was applied to a representative sample of people and a group of musical experts.
This item is licensed under a Creative Commons License