Date of Award

2019

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Dr. Jingsheng Xu

Abstract

With the growth of digital audio data, various and fast access to music data is strongly desired, especially for large music databases. A more natural way to retrieve a song from a database will be to hum to the tune. To relate and compare musical pieces is a very complex task. Musical compositions usually collapse multiple information sources and complex, multifaceted interactions established between parts. Despite such degrees of complexity, humans are outstandingly good at performing individual musical judgments with little conscious effort, while a computer cannot efficiently achieve this task. In this work, we focus on one such task: music information retrieval using content-based input such as users’ hum to tune a music piece. Query by humming (QBH) is another content-based retrieval method for Music Information Retrieval in an extensive music database. This method is a form of audio to audio mapping of events in one recording to the similar events in the other recording. In particular, we adopt computational approach information provided by the audio signal. We propose a system for music retrieval that matches a user humming to the best song in an audio database. We developed an algorithm to map the difference in two musical pieces in time series domain based on dynamic programming and longest subsequence. We explored alternative tunes for user query input, which might be off-tune, thereby achieving high accuracy on query identification from a music database. However, the accuracy of the result is highly dependent on the query quality and closeness to the actual song.

Download

Share

COinS