Abstract / Excerpt:
The initial idea of this study is to search a keyword/ phrase in an audio file using audio mining, display its filename and the location of the keyword/ phrase in the audio file to the user. This is to make the audio mining technology since it hasn't been used in this way before. However, due to certain circumstances, the initial study was not pursued. Therefore, the proponents decided to shift the idea into something similar to the desired output which is to display the title of the song, or its filename to the user after a search has been done. Lyric fetching was used. This is to make it possible for the user to search the lyrics of the audio files given the search keyword/ phrase. Three APIs were used to fetch lyrics from a wide range of database in each respective host lyric fetching website. The user is allowed to choose which lyric result is desired since it does an exhaustive search, which means that it makes use of all the lyric fetch results. After fetching the lyrics, the system saves it in a local database. The searching functionality makes use of the Boyer-Moore algorithm. Search keyword/ phrase is matched to the lyrics of an audio file. The audio file with the most number of search results is the most probable desired audio file to be searched. Further study of the audio mining technology is also recommended.
|Source Institution||Ateneo de Davao University|
|Authors||Morny Bernie G. Bolcan, John Ray H. Piedraverde|
|Place of Publication||Davao City|
|Original Publication Date||March 1, 2010|