Next generation radio streaming service towards combination of features of traditional radio and a music streaming service
Citation & Export
Hide
Simple citation
Chawla, Tanuj.
Next generation radio streaming service towards combination of features of traditional radio and a music streaming service. Retrieved from
https://doi.org/doi:10.7282/T31838XQExport
Description
TitleNext generation radio streaming service towards combination of features of traditional radio and a music streaming service
Date Created2017
Other Date2017-01 (degree)
Extent1 online resource (x, 43 p. : ill.)
Description“Our current way of listening to music is actually harder than it is to listen to the radio, which is still strangely popular among Americans. This suggests that a lot of people are looking for not only music that they like, but an easy listening experience that allows them to simply consume and not create.” It doesn’t matter how many million songs are offered by one’s streaming music service, people still find it hard to decide each time that what they want to listen. It can be downright stressful to choose the right music at the right moment — and stress is anything but convenient. Furthermore, think about doing this by interacting with the phone screen while driving 65mph on a highway or running 10mph on a treadmill. It can soon turn quite dangerous, resulting in a need for a much simpler user interface for these applications. TapTap Radio - our proposed next generation radio streaming service offers a combination of features of traditional radio and a streaming music service. Given a very large collection of music (millions of songs) it combines user control of music players with randomness characteristic of traditional radio. TapTap Radio offers seamless discovery of new music together with the ability to “focus” and “freeze” it on the music which the user likes. The interface for TapTap Radio allows for simple interactions (tapping on the smart phone screen) for a user to change the station which plays a stream of songs defined by a set of predicates. Users can create customized radio stations by tweaking song parameters and the number of such stations is unlimited just as the number of combinations of these parameters such as genre, style, popularity, release date, record type etc. Stations are more than playlists since they are dynamically defined and carry different content as the underlying database of songs (Spotify, Rovi Music and Gramusik) is updated regularly. Users can move from station to station, do a voice search, move within a station unlike radio, repeat the song or get more songs from the playing artist, play similar songs, all just with the combination of taps on the screen (analogous to Morse Code but simpler). The tapping interface is not just limited to music streaming apps but can also be extended to a whole bunch of other applications, for example, controlling IoT devices, all media applications, any application with a set of controls required to function with limited ability to interact. We also introduce two learning algorithms which unlike Pandora and other services, learns user preferences with minimal user intervention by exploiting the routes that user navigates through the app. Performance analysis is also done for both the algorithms. Based on these algorithms, recommendations can be made to the user and new stations can be generated for them as a result of their acceptance. Hence, TapTap Radio would enable serious music enthusiasts to still be able to play what they like when they want it. But the whole bunch of naive listeners out there will simply find themselves magically listening to just the right songs they love. And for those sudden mood swings we have got them covered with a set of always playing radio stations and another set of radio stations which are based on day of week and time of day.
NoteM.S.
NoteIncludes bibliographical references
Noteby Tanuj Chawla
Genretheses, ETD graduate
Languageeng
CollectionGraduate School - New Brunswick Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Statistical Profile