IntegratedRecognition: A Theoretically More Efficient Approach to Cached Music Recognition

Restricted (Penn State Only)
Author:
Jiang, Xiaoxuan
Graduate Program:
Computer Science and Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
November 06, 2018
Committee Members:
  • Mahmut Taylan Kandemir, Thesis Advisor
Keywords:
  • Integrated Music Recognition
  • Locality Recognition
Abstract:
Most existing music recognition applications are independent from music applications. In this paper we discuss a theoretically more efficient approach, called IntegratedRecognition, to recognize music from a pool based on a cached music listening habits. Instead of being independent from music applications, IntegratedRecognition is designed to be implemented as an add-on function to the music application. This will help the developers get users’ music listening preferences necessary to narrow down the scope of landmark comparison. At the end, future research directions on embedded music recognition will be discussed.