Tunebot

Researchers

Bryan Pardo, Mark Cartwright, Arefin Huq, Denis Lebel, David Little, Zafa Rafii, Jinyu Han

Overview

Tunebot was a query-by-humming search engine that was online from 2008–2016. It let you find the music you're looking for by singing a bit of it. In response to your query it returned a ranked list of songs you can play. The songs were linked to Amazon.com, where you could purchase the desired music.

As one of the researchers and developers of Tunebot, I developed aspects of the front-end and back-end, as well as investigated personalization and methods for crowdsourcing the database of music.

Related Papers

[pdf] Cartwright, M., Pardo, B. Building a Music Search Database Using Human Computation. In Proceedings of the Sound and Music Computing Conference (SMC), 2012.

[pdf] [poster] Cartwright, M., Rafii, Z., Han, J., Pardo, B. Making Searchable Melodies: Human vs. Machine. In Proceedings of the 2011 AAAI Workshop on Human Computation (HCOMP), 2011.

[pdf] Huq, A., Cartwright, M., Pardo, B. Crowdsourcing a Real-World On-Line Query By Humming System. In Proceedings of the Sound and Music Computing Conference (SMC), 2010.

Datasets

Tunebot Dataset