Off-grid Fundamental Frequency Estimation
(2018) In IEEE/ACM Transactions on Audio, Speech, and Language Processing 26(2). p.296-303- Abstract
In this paper, we propose a gridless method for estimating an unknown number of fundamental frequencies. Starting with a conventional dictionary matrix, containing sets of candidate fundamental frequencies and their corresponding harmonics, a non-convex log-sum cost function is formed such that it imposes the harmonic structure and treats every fundamental frequency in the dictionary as a parameter. The cost function is iteratively decreased by minimizing a surrogate function, and, in each iteration, the fundamental frequencies are refined, whereas redundant parameters are omitted from the dictionary. The proposed method is tested on both real and simulated data, showing its preferred performance as compared to other state-of-the-art... (More)
In this paper, we propose a gridless method for estimating an unknown number of fundamental frequencies. Starting with a conventional dictionary matrix, containing sets of candidate fundamental frequencies and their corresponding harmonics, a non-convex log-sum cost function is formed such that it imposes the harmonic structure and treats every fundamental frequency in the dictionary as a parameter. The cost function is iteratively decreased by minimizing a surrogate function, and, in each iteration, the fundamental frequencies are refined, whereas redundant parameters are omitted from the dictionary. The proposed method is tested on both real and simulated data, showing its preferred performance as compared to other state-of-the-art multi-pitch estimators.
(Less)
- author
- Sward, Johan LU ; Li, Hongbin and Jakobsson, Andreas LU
- organization
- publishing date
- 2018-02
- type
- Contribution to journal
- publication status
- published
- subject
- in
- IEEE/ACM Transactions on Audio, Speech, and Language Processing
- volume
- 26
- issue
- 2
- pages
- 296 - 303
- publisher
- Piscataway, NJ : Institute of Electrical and Electronics Engineers
- external identifiers
-
- wos:000418297800007
- scopus:85035785863
- ISSN
- 2329-9290
- DOI
- 10.1109/TASLP.2017.2775800
- language
- English
- LU publication?
- yes
- id
- 9f3ffb30-5a15-4d96-97ba-98f1e53946b0
- date added to LUP
- 2017-12-12 13:21:35
- date last changed
- 2024-06-10 06:07:52
@article{9f3ffb30-5a15-4d96-97ba-98f1e53946b0, abstract = {{<p>In this paper, we propose a gridless method for estimating an unknown number of fundamental frequencies. Starting with a conventional dictionary matrix, containing sets of candidate fundamental frequencies and their corresponding harmonics, a non-convex log-sum cost function is formed such that it imposes the harmonic structure and treats every fundamental frequency in the dictionary as a parameter. The cost function is iteratively decreased by minimizing a surrogate function, and, in each iteration, the fundamental frequencies are refined, whereas redundant parameters are omitted from the dictionary. The proposed method is tested on both real and simulated data, showing its preferred performance as compared to other state-of-the-art multi-pitch estimators.</p>}}, author = {{Sward, Johan and Li, Hongbin and Jakobsson, Andreas}}, issn = {{2329-9290}}, language = {{eng}}, number = {{2}}, pages = {{296--303}}, publisher = {{Piscataway, NJ : Institute of Electrical and Electronics Engineers}}, series = {{IEEE/ACM Transactions on Audio, Speech, and Language Processing}}, title = {{Off-grid Fundamental Frequency Estimation}}, url = {{http://dx.doi.org/10.1109/TASLP.2017.2775800}}, doi = {{10.1109/TASLP.2017.2775800}}, volume = {{26}}, year = {{2018}}, }