Music Version Retrieval from YouTube: How to Formulate Effective Search Queries?
S. Hachmeier, R. Jäschke, and H. Saadatdoorabi. Proceedings of the Conference on ``Lernen, Wissen, Daten, Analysen'', 3341, page 213--226. Aachen, (2022)
Abstract
Various versions of musical works are published on YouTube, such as remixes or reaction videos. While some research has focused on tasks like audio-based version identification of these videos, it is still unclear how to effectively retrieve a large amount of relevant versions with textual queries. In this paper, we formulate search queries with YouTube search suggestions, evaluate these based on multiple dimensions and compute optimal ranks of queries on work-level. We show that queries containing the artist string retrieve results with higher relevance, but have higher overlaps. Additionally, we demonstrate that the amount of reasonable queries can be increased by applying frequently suggested expansions to works which tend to contextualize queries to the music domain.
%0 Conference Paper
%1 hachmeier2022music
%A Hachmeier, Simon
%A Jäschke, Robert
%A Saadatdoorabi, Hadi
%B Proceedings of the Conference on ``Lernen, Wissen, Daten, Analysen''
%C Aachen
%D 2022
%E Reuss, Pascal
%E Eisenstadt, Viktor
%E Schönborn, Jakob
%E Schäfer, Jero
%K 2022 cover information lwda mir music myown retrieval song youtube
%N 3341
%P 213--226
%T Music Version Retrieval from YouTube: How to Formulate Effective Search Queries?
%U https://ceur-ws.org/Vol-3341/WM-LWDA_2022_CRC_7142.pdf
%X Various versions of musical works are published on YouTube, such as remixes or reaction videos. While some research has focused on tasks like audio-based version identification of these videos, it is still unclear how to effectively retrieve a large amount of relevant versions with textual queries. In this paper, we formulate search queries with YouTube search suggestions, evaluate these based on multiple dimensions and compute optimal ranks of queries on work-level. We show that queries containing the artist string retrieve results with higher relevance, but have higher overlaps. Additionally, we demonstrate that the amount of reasonable queries can be increased by applying frequently suggested expansions to works which tend to contextualize queries to the music domain.
@inproceedings{hachmeier2022music,
abstract = {Various versions of musical works are published on YouTube, such as remixes or reaction videos. While some research has focused on tasks like audio-based version identification of these videos, it is still unclear how to effectively retrieve a large amount of relevant versions with textual queries. In this paper, we formulate search queries with YouTube search suggestions, evaluate these based on multiple dimensions and compute optimal ranks of queries on work-level. We show that queries containing the artist string retrieve results with higher relevance, but have higher overlaps. Additionally, we demonstrate that the amount of reasonable queries can be increased by applying frequently suggested expansions to works which tend to contextualize queries to the music domain.},
added-at = {2022-09-19T08:31:30.000+0200},
address = {Aachen},
author = {Hachmeier, Simon and Jäschke, Robert and Saadatdoorabi, Hadi},
biburl = {https://www.bibsonomy.org/bibtex/2f4ec17cc6a1076b8664114f976e0be0e/jaeschke},
booktitle = {Proceedings of the Conference on ``Lernen, Wissen, Daten, Analysen''},
editor = {Reuss, Pascal and Eisenstadt, Viktor and Schönborn, Jakob and Schäfer, Jero},
interhash = {ec757529f36f515ad8ba9dd8a1233554},
intrahash = {f4ec17cc6a1076b8664114f976e0be0e},
issn = {1613-0073},
keywords = {2022 cover information lwda mir music myown retrieval song youtube},
number = 3341,
pages = {213--226},
series = {CEUR Workshop Proceedings},
timestamp = {2024-02-13T14:26:19.000+0100},
title = {Music Version Retrieval from YouTube: How to Formulate Effective Search Queries?},
url = {https://ceur-ws.org/Vol-3341/WM-LWDA_2022_CRC_7142.pdf},
year = 2022
}