An Adaptive Penalty Approach to Multi-Pitch Estimation

Ted Kronvall, Filip Elvander, Stefan Ingi Adalbjörnsson, Andreas Jakobsson

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review

188 Downloads (Pure)


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.
Original languageEnglish
Title of host publication Signal Processing Conference (EUSIPCO), 2015 23rd European
Number of pages5
ISBN (Electronic)978-0-9928-6263-3
Publication statusPublished - 2015 Dec 28
Event23rd European Signal Processing Conference, 2015 - Nice, France
Duration: 2015 Aug 312015 Sept 4
Conference number: 23

Publication series

NameEuropean Signal Processing Conference (EUSIPCO)
ISSN (Electronic)2076-1465


Conference23rd European Signal Processing Conference, 2015
Abbreviated titleEUSIPCO

Subject classification (UKÄ)

  • Probability Theory and Statistics
  • Signal Processing


  • multi-pitch estimation
  • block sparsity
  • adaptive sparse penalty
  • total variation
  • ADMM


Dive into the research topics of 'An Adaptive Penalty Approach to Multi-Pitch Estimation'. Together they form a unique fingerprint.

Cite this