Advanced

An Adaptive Penalty Approach to Multi-Pitch Estimation

Kronvall, Ted LU ; Elvander, Filip LU ; Adalbjörnsson, Stefan Ingi LU and Jakobsson, Andreas LU (2015) 23rd European Signal Processing Conference, 2015 In Signal Processing Conference (EUSIPCO), 2015 23rd European
Abstract
This work treats multi-pitch estimation, and in particular the common misclassification issue wherein the pitch at half of the true fundamental frequency, here referred to as a sub-octave, is chosen instead of the true pitch. Extending on current methods which use an extension of the Group LASSO for pitch estimation, this work introduces an adaptive total variation penalty, which both enforce group- and block sparsity, and deal with errors due to sub-octaves. The method is shown to outperform current state-of-the-art sparse methods, where the model orders are unknown, while also requiring fewer tuning parameters than these. The method is also shown to outperform several conventional pitch estimation methods, even when these are virtued... (More)
This work treats multi-pitch estimation, and in particular the common misclassification issue wherein the pitch at half of the true fundamental frequency, here referred to as a sub-octave, is chosen instead of the true pitch. Extending on current methods which use an extension of the Group LASSO for pitch estimation, this work introduces an adaptive total variation penalty, which both enforce group- and block sparsity, and deal with errors due to sub-octaves. The method is shown to outperform current state-of-the-art sparse methods, where the model orders are unknown, while also requiring fewer tuning parameters than these. The method is also shown to outperform several conventional pitch estimation methods, even when these are virtued with oracle model orders. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
multi-pitch estimation, block sparsity, adaptive sparse penalty, total variation, ADMM
in
Signal Processing Conference (EUSIPCO), 2015 23rd European
pages
5 pages
publisher
EURASIP
conference name
23rd European Signal Processing Conference, 2015
external identifiers
  • Scopus:84963959097
ISSN
2076-1465
DOI
10.1109/EUSIPCO.2015.7362339
language
English
LU publication?
yes
id
a9e15a0b-7268-4ec6-b420-c5b97ebfb2ac (old id 8046477)
date added to LUP
2015-10-08 13:33:16
date last changed
2017-01-01 08:32:44
@inproceedings{a9e15a0b-7268-4ec6-b420-c5b97ebfb2ac,
  abstract     = {This work treats multi-pitch estimation, and in particular the common misclassification issue wherein the pitch at half of the true fundamental frequency, here referred to as a sub-octave, is chosen instead of the true pitch. Extending on current methods which use an extension of the Group LASSO for pitch estimation, this work introduces an adaptive total variation penalty, which both enforce group- and block sparsity, and deal with errors due to sub-octaves. The method is shown to outperform current state-of-the-art sparse methods, where the model orders are unknown, while also requiring fewer tuning parameters than these. The method is also shown to outperform several conventional pitch estimation methods, even when these are virtued with oracle model orders.},
  author       = {Kronvall, Ted and Elvander, Filip and Adalbjörnsson, Stefan Ingi and Jakobsson, Andreas},
  booktitle    = { Signal Processing Conference (EUSIPCO), 2015 23rd European},
  issn         = {2076-1465},
  keyword      = {multi-pitch estimation,block sparsity,adaptive sparse penalty,total variation,ADMM},
  language     = {eng},
  month        = {12},
  pages        = {5},
  publisher    = {EURASIP},
  title        = {An Adaptive Penalty Approach to Multi-Pitch Estimation},
  url          = {http://dx.doi.org/ 10.1109/EUSIPCO.2015.7362339},
  year         = {2015},
}