Joint Fundamental Frequency and Order Estimation using Optimal Filtering
Christensen, Mads ; Jensen, Jesper ; Jakobsson, Andreas LU
and Jensen, Sören
(2009)
17th European Signal Processing Conference
- Abstract
- In this paper, two optimal filter designs for fundamental frequency
estimation are presented with the first being based on a filterbank
and the second on a single filter. We show that the two designs are
related in a simple manner and result in the same residual when used
for cancelling out the harmonics. We propose to use this residual
for estimating the number of harmonics by combining a noise variance
estimate with an order dependent penalty term. This leads to
a joint estimator of the fundamental frequency and the order based
on the same criterion. Via Monte Carlo simulations, the estimator is
demonstrated to have good performance in terms of the... (More) - In this paper, two optimal filter designs for fundamental frequency
estimation are presented with the first being based on a filterbank
and the second on a single filter. We show that the two designs are
related in a simple manner and result in the same residual when used
for cancelling out the harmonics. We propose to use this residual
for estimating the number of harmonics by combining a noise variance
estimate with an order dependent penalty term. This leads to
a joint estimator of the fundamental frequency and the order based
on the same criterion. Via Monte Carlo simulations, the estimator is
demonstrated to have good performance in terms of the percentage
of correctly estimated orders. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/1397940
- author
- Christensen, Mads
; Jensen, Jesper
; Jakobsson, Andreas
LU
and Jensen, Sören
- organization
- publishing date
- 2009
- type
- Contribution to conference
- publication status
- in press
- subject
- keywords
- Acoustic signal analysis, spectral analysis, frequency estimation
- conference name
- 17th European Signal Processing Conference
- conference dates
- 2009-08-24 - 2009-08-28
- language
- English
- LU publication?
- yes
- id
- e1895e56-0a1a-4965-b1fd-d46d40d49e61 (old id 1397940)
- alternative location
- http://kom.aau.dk/~mgc/publications/icassp2009.pdf
- date added to LUP
- 2016-04-04 13:19:50
- date last changed
- 2018-11-21 21:13:17
@misc{e1895e56-0a1a-4965-b1fd-d46d40d49e61,
abstract = {{In this paper, two optimal filter designs for fundamental frequency<br/><br>
estimation are presented with the first being based on a filterbank<br/><br>
and the second on a single filter. We show that the two designs are<br/><br>
related in a simple manner and result in the same residual when used<br/><br>
for cancelling out the harmonics. We propose to use this residual<br/><br>
for estimating the number of harmonics by combining a noise variance<br/><br>
estimate with an order dependent penalty term. This leads to<br/><br>
a joint estimator of the fundamental frequency and the order based<br/><br>
on the same criterion. Via Monte Carlo simulations, the estimator is<br/><br>
demonstrated to have good performance in terms of the percentage<br/><br>
of correctly estimated orders.}},
author = {{Christensen, Mads and Jensen, Jesper and Jakobsson, Andreas and Jensen, Sören}},
keywords = {{Acoustic signal analysis; spectral analysis; frequency
estimation}},
language = {{eng}},
title = {{Joint Fundamental Frequency and Order Estimation using Optimal Filtering}},
url = {{http://kom.aau.dk/~mgc/publications/icassp2009.pdf}},
year = {{2009}},
}