A Classification Approach to Melody Transcription
Year: 2005
Type: article
Abstract: Melodies provide an important conceptual summarization of polyphonic audio. The extraction of melodic content has practical applications ranging from content-based audio retrieval to the analysis of musical structure. In contrast to previous transcription systems based on a model of the harmonic (or periodic) structure of musical pitches, we present a classification-based system for performing automatic melody transcription that makes no assumptions beyond what is learned from its training data. We evaluate the success of our algorithm by predicting the melody of the ISMIR 2004 Melody Competition evaluation set and on newly-generated test data. We show that a Support Vector Machine melodic classifier produces results comparable to state of the art model-based transcription systems.
DOI: 10.7916/d81z4dt7
Authors: Poliner, Graham E., Ellis, Daniel P. W.
Institution: Columbia University
Language: English
Topic: Music and Audio Processing
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: National Science Foundation
OpenAlex ID: W1521419984
