Raitmo: generador musical para apoyar la composición musical basado en machine learning
Improv RNN is a tool based on Google?s Magenta library which can generate new melodies based on a deep learning model, but there is a lack in documentation about its use and how to generate the dataset for the training step, which makes it difficult to use for users with low or none experience in so...
- Autores:
-
Patarroyo Devia, José Leonardo
Yepes Bonilla, álvaro Alfonso
- Tipo de recurso:
- Trabajo de grado de pregrado
- Fecha de publicación:
- 2019
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/51464
- Acceso en línea:
- http://hdl.handle.net/1992/51464
- Palabra clave:
- Aprendizaje automático (Inteligencia artificial)
Composición musical
Ingeniería
- Rights
- openAccess
- License
- http://creativecommons.org/licenses/by-nc-nd/4.0/
| Summary: | Improv RNN is a tool based on Google?s Magenta library which can generate new melodies based on a deep learning model, but there is a lack in documentation about its use and how to generate the dataset for the training step, which makes it difficult to use for users with low or none experience in software programming. Given the above, the current project aimed to solve the problem by two main results: development of an application that uses a model of Improv RNN trained with famous reggaeton melodies to generate new melodies, and an interactive web blog that explains step by step the development of the model used in the application including theoretical and practical considerations to understand it. |
|---|