Abstract
With the considerable progress of information technologies, various music recommendation services, including traditional music radio stations and Internet radio services, are available on the Internet. Most music programs have their own moods or atmospheres, depending on broadcasting time, target listeners, and program purpose. Essentially, a significant portion of this atmosphere depends on the music playlist of the program. This indicates that each radio program has some consistent tendency or trend in its music selection. In this paper, we propose a music radio program cloning technique, called SmartRadio based on various musical features and context information for music recommendations of the same quality as real radio programs. We implemented a prototype SmartRadio system and showed that SmartRadio can produce satisfactory results through our experiments.
Original language | English |
---|---|
Pages (from-to) | 2701-2709 |
Number of pages | 9 |
Journal | Information (Japan) |
Volume | 16 |
Issue number | 4 |
Publication status | Published - 2013 Apr |
Keywords
- Music feature extraction
- Music station
- Signal processing
ASJC Scopus subject areas
- Information Systems