00000ctm a22000004a 4500 UP-99796217611172895 Buklod 20230215105418.0 m |o d | ta 140329s xx d r |||| u| (iLib)UPD-00214606202 DENGII eng DMLUC LG 993.5 2011 E64 L34 Laguna, Ann Franchesca Real-time karaoke grading visualization interface using modal distribution for mobile devices Ann Franchesca Laguna. 2011. viii, 61 leaves ill. (some col.) + 1 DVD-ROM (4 3/4 in.) Undergraduate student project (B.S. Computer Engineering)--University of the Philippines, Diliman. The objective of the researcher is to investigate the viability of porting a real-time karaoke grader from a PC to a mobile device, particularly the Samsung GT-i5700 Galaxy Spica running on an 800MHz Processor with an ARM1176JZS CPU Core. The software is subdivided into different modules namely: the MIDI Parser Module, Modal Distribution Frequency and Amplitude Module, the Scorer Module and the Graphical User Interface Module. The karaoker grader has two inputs: the MIDI file and the voice input of the user. The MIDI file contains melody, lyrics and accompaniment tracks. The MIDI Parser Module takes a MIDI file and parses it to obtain the melody notes to be used as the basis for grading. A PC-based implementation and an Android based implementation of the MIDI Parser were both developed. Modal Distribution was used to estimate the amplitude and frequency of the singer's input voice. These estimates are converted into tempo and pitch scores with a resolution of 10 ms for tempo and 3.25Hz for pitch. Two versions of the karaoke grader were developed: online and offline. The online version has missed notes due to the computing demands of the modal distribution. The computing power of the mobile device can only support the offline version with an accuracy of 99.52% for the pitch and 98.73% for the tempo. Computer sound processing. Music Data processing. Mobile computing. Modal distribution analysis. Karaoke grading system. Thesis FI UP Thesis UPD DENG-II LG 993.5 2011 E64 L34 Thesis