Music identification

The digitization of music and increase in streaming platforms has caused an information overload. Therefore, it is important to retrieve critical musical data such as pitch, tempo and timbre. This could lead to several commercially viable applications such as music recommendation, classification, co...

Full description

Saved in:
Bibliographic Details
Main Author: Manikandan, Varsha
Other Authors: Anamitra Makur
Format: Final Year Project
Language:English
Published: 2018
Subjects:
Online Access:http://hdl.handle.net/10356/75440
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:The digitization of music and increase in streaming platforms has caused an information overload. Therefore, it is important to retrieve critical musical data such as pitch, tempo and timbre. This could lead to several commercially viable applications such as music recommendation, classification, cover song detection, etc. The objective of this project was to develop an application that could recognize a song from a database despite adjustments in tempo and frequency. This was done with the help of MATLAB software. The music recognition was done from a set database, containing songs played on a flute. Tempo and Pitch adjustments were made to the songs and passed as input query. The application could successfully detect changes in beat and tempo and match the songs from the database. The results were represented through a graphical user interface, that lets the user upload their query. The GUI finds the best match and plots the two spectrograms.