An Adaptive Penalty Approach to Multi-Pitch Estimation
(2015) 23rd European Signal Processing Conference, 2015 In European Signal Processing Conference (EUSIPCO)- 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:
https://lup.lub.lu.se/record/8046477
- author
- Kronvall, Ted LU ; Elvander, Filip LU ; Adalbjörnsson, Stefan Ingi LU and Jakobsson, Andreas LU
- organization
- publishing date
- 2015-12-28
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- multi-pitch estimation, block sparsity, adaptive sparse penalty, total variation, ADMM
- host publication
- Signal Processing Conference (EUSIPCO), 2015 23rd European
- series title
- European Signal Processing Conference (EUSIPCO)
- pages
- 5 pages
- publisher
- EURASIP
- conference name
- 23rd European Signal Processing Conference, 2015
- conference location
- Nice, France
- conference dates
- 2015-08-31 - 2015-09-04
- external identifiers
-
- scopus:84963959097
- ISSN
- 2076-1465
- ISBN
- 978-0-9928-6263-3
- DOI
- 10.1109/EUSIPCO.2015.7362339
- language
- English
- LU publication?
- yes
- id
- a9e15a0b-7268-4ec6-b420-c5b97ebfb2ac (old id 8046477)
- date added to LUP
- 2016-04-04 13:38:48
- date last changed
- 2022-01-30 05:41:33
@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}}, isbn = {{978-0-9928-6263-3}}, issn = {{2076-1465}}, keywords = {{multi-pitch estimation; block sparsity; adaptive sparse penalty; total variation; ADMM}}, language = {{eng}}, month = {{12}}, publisher = {{EURASIP}}, series = {{European Signal Processing Conference (EUSIPCO)}}, title = {{An Adaptive Penalty Approach to Multi-Pitch Estimation}}, url = {{https://lup.lub.lu.se/search/files/11421050/8046478.pdf}}, doi = {{10.1109/EUSIPCO.2015.7362339}}, year = {{2015}}, }