Open AccessTimbre Comparison in Note Tracking from Onset, Frames and Pitch Estimation
Author(s) -
Carlos Hernández Oliván,
Ignacio Zay Pinilla,
José Ramón Beltrán Blázquez
Publication year - 2020
Publication title -
jornada de jóvenes investigadores del i3a
Language(s) - English
Resource type - Journals
ISSN - 2341-4790
DOI - 10.26754/jjii3a.4872
Subject(s) - timbre , tracking (education) , computer science , speech recognition , field (mathematics) , transcription (linguistics) , artificial intelligence , computer vision , mathematics , psychology , musical , art , visual arts , linguistics , pedagogy , philosophy , pure mathematics
Note Tracking (NT) is a subtask of Automatic Music Transcription (AMT) which is a critical problem in the field of Music Information Retrieval (MIR). The aim of this work is to compare the performance of two models, one for onsets and frames prediction and another one with pitch detection and a note tracking algorithm in order to study the behaviour of different timbres and families of instruments in note tracking subtasks.