Detail publikace

Comparison of fundamental frequency detection methods and introducing self-repairing algorithm for musical applications

Originální název

Comparison of fundamental frequency detection methods and introducing self-repairing algorithm for musical applications

Anglický název

Comparison of fundamental frequency detection methods and introducing self-repairing algorithm for musical applications

Jazyk

en

Originální abstrakt

This paper presents the comparison of five commonly used methods for fundamental frequency detection in speech signal, exactly in vocal and melodic instrument signals. The efficiency of chosen method is verified on known set of musical notes performed by bass clarinet. The highest efficiency in fundamental frequency detection was reached by AutoCorrelation (ACF) and Modified AutoCorrelation (MACF) functions. Self-repairing algorithm is also described in this paper and it can be defined as a useful tool for correction of inaccurately found fundamental frequencies related to relevant musical notes. For correct pitch detection and self-repairing algorithm function, as the most appropriate segment length can be set as a half value of the shortest known musical note in analysed signal. Due to high efficiency and low computational preformance, the combination of ACF, MACF respectively, with self-repairing algorithm supplemented by some fundamental frequency changing method can be used as an effective almost real-time tool for tuning and other musical applications.

Anglický abstrakt

This paper presents the comparison of five commonly used methods for fundamental frequency detection in speech signal, exactly in vocal and melodic instrument signals. The efficiency of chosen method is verified on known set of musical notes performed by bass clarinet. The highest efficiency in fundamental frequency detection was reached by AutoCorrelation (ACF) and Modified AutoCorrelation (MACF) functions. Self-repairing algorithm is also described in this paper and it can be defined as a useful tool for correction of inaccurately found fundamental frequencies related to relevant musical notes. For correct pitch detection and self-repairing algorithm function, as the most appropriate segment length can be set as a half value of the shortest known musical note in analysed signal. Due to high efficiency and low computational preformance, the combination of ACF, MACF respectively, with self-repairing algorithm supplemented by some fundamental frequency changing method can be used as an effective almost real-time tool for tuning and other musical applications.

BibTex


@inproceedings{BUT114246,
  author="Miroslav {Staněk} and Tomáš {Smatana}",
  title="Comparison of fundamental frequency detection methods and introducing self-repairing algorithm for musical applications",
  annote="This paper presents the comparison of five
commonly used methods for fundamental frequency detection in
speech signal, exactly in vocal and melodic instrument signals.
The efficiency of chosen method is verified on known set of
musical notes performed by bass clarinet. The highest efficiency
in fundamental frequency detection was reached by
AutoCorrelation (ACF) and Modified AutoCorrelation (MACF)
functions. Self-repairing algorithm is also described in this paper
and it can be defined as a useful tool for correction of
inaccurately found fundamental frequencies related to relevant
musical notes. For correct pitch detection and self-repairing
algorithm function, as the most appropriate segment length can
be set as a half value of the shortest known musical note in
analysed signal. Due to high efficiency and low computational
preformance, the combination of ACF, MACF respectively, with
self-repairing algorithm supplemented by some fundamental
frequency changing method can be used as an effective almost
real-time tool for tuning and other musical applications.
",
  booktitle="Proceedings of 25th International Conference Radioelektronika",
  chapter="114246",
  howpublished="print",
  year="2015",
  month="april",
  pages="217--221",
  type="conference paper"
}