TY - GEN
T1 - Query by Humming for Song Identification Using Voice Isolation
AU - Alfaro-Paredes, Edwin
AU - Alfaro-Carrasco, Leonardo
AU - Ugarte, Willy
N1 - Publisher Copyright:
© 2021, Springer Nature Switzerland AG.
PY - 2021
Y1 - 2021
N2 - There are some methods for searching in large music databases, like searching by song name or artist name. However, in some cases these methods are not enough. For instance, a person might not remember the name of a song, but might remember its melody. In Music Information Retrieval, there is a task called Query-By-Humming, which allows retrieving a rank of songs that are similar to an audio humming. In this research, we propose the use of vocal isolation methods to improve query-by-humming systems. To achieve this, different configurations of Query-by-Humming systems were tested to analyze the results and determine in which cases our proposal works better. The results showed that vocal isolation improves the performance of Query-by-Humming systems when the music collection consists of modern songs.
AB - There are some methods for searching in large music databases, like searching by song name or artist name. However, in some cases these methods are not enough. For instance, a person might not remember the name of a song, but might remember its melody. In Music Information Retrieval, there is a task called Query-By-Humming, which allows retrieving a rank of songs that are similar to an audio humming. In this research, we propose the use of vocal isolation methods to improve query-by-humming systems. To achieve this, different configurations of Query-by-Humming systems were tested to analyze the results and determine in which cases our proposal works better. The results showed that vocal isolation improves the performance of Query-by-Humming systems when the music collection consists of modern songs.
KW - Melody extraction
KW - Music information retrieval
KW - Music similarity
KW - Query-by-Humming
UR - https://www.scopus.com/pages/publications/85112695388
U2 - 10.1007/978-3-030-79463-7_27
DO - 10.1007/978-3-030-79463-7_27
M3 - Contribución a la conferencia
AN - SCOPUS:85112695388
SN - 9783030794620
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 323
EP - 334
BT - Advances and Trends in Artificial Intelligence. From Theory to Practice - 34th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2021, Proceedings
A2 - Fujita, Hamido
A2 - Selamat, Ali
A2 - Lin, Jerry Chun-Wei
A2 - Ali, Moonis
PB - Springer Science and Business Media Deutschland GmbH
T2 - 34th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2021
Y2 - 26 July 2021 through 29 July 2021
ER -