Games based on human computation are a valuable tool for collecting semantic information about images. We show how to transfer this idea into the music domain in order to collect high-quality semantic data for songs. We present Listen Game, a online, multiplayer game that measures the semantic relationship between music and words. In the normal mode, a player sees a list of semantically related words (e.g., ‘Instruments’, ‘Emotions ’ ‘Usages’) and is asked to pick the best and worst word to describe a song. In the freestyle mode, a user is asked to suggest a new word that describes the music. Each player receives realtime feedback on the agreement amongst all players. We show that we can use the data collected during a twoweek pilot study of Listen Game to learn a supervised multiclass labeling (SML) model which can annotate a novel song with meaningful words and retrieve relevant songs from a database of audio content.
%0 Conference Paper
%1 Turnbull07agame-based
%A Turnbull, Douglas
%A Liu, Ruoran
%A Barrington, Luke
%A Lanckriet, Gert
%B In 8th International Conference on Music Information Retrieval (ISMIR
%D 2007
%K annotation game music semantic
%T A game-based approach for collecting semantic annotations of music
%X Games based on human computation are a valuable tool for collecting semantic information about images. We show how to transfer this idea into the music domain in order to collect high-quality semantic data for songs. We present Listen Game, a online, multiplayer game that measures the semantic relationship between music and words. In the normal mode, a player sees a list of semantically related words (e.g., ‘Instruments’, ‘Emotions ’ ‘Usages’) and is asked to pick the best and worst word to describe a song. In the freestyle mode, a user is asked to suggest a new word that describes the music. Each player receives realtime feedback on the agreement amongst all players. We show that we can use the data collected during a twoweek pilot study of Listen Game to learn a supervised multiclass labeling (SML) model which can annotate a novel song with meaningful words and retrieve relevant songs from a database of audio content.
@inproceedings{Turnbull07agame-based,
abstract = {Games based on human computation are a valuable tool for collecting semantic information about images. We show how to transfer this idea into the music domain in order to collect high-quality semantic data for songs. We present Listen Game, a online, multiplayer game that measures the semantic relationship between music and words. In the normal mode, a player sees a list of semantically related words (e.g., ‘Instruments’, ‘Emotions ’ ‘Usages’) and is asked to pick the best and worst word to describe a song. In the freestyle mode, a user is asked to suggest a new word that describes the music. Each player receives realtime feedback on the agreement amongst all players. We show that we can use the data collected during a twoweek pilot study of Listen Game to learn a supervised multiclass labeling (SML) model which can annotate a novel song with meaningful words and retrieve relevant songs from a database of audio content.},
added-at = {2008-10-15T13:59:44.000+0200},
author = {Turnbull, Douglas and Liu, Ruoran and Barrington, Luke and Lanckriet, Gert},
biburl = {https://www.bibsonomy.org/bibtex/2daa312cc50ee2e53a48fe78f88586a8f/jaeschke},
booktitle = {In 8th International Conference on Music Information Retrieval (ISMIR},
interhash = {84dbb47780598f72fb33e7e1e431fdb9},
intrahash = {daa312cc50ee2e53a48fe78f88586a8f},
keywords = {annotation game music semantic},
timestamp = {2014-07-28T15:57:31.000+0200},
title = {A game-based approach for collecting semantic annotations of music},
year = 2007
}