Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Computer identification of musical instruments using pattern recognition with cepstral coefficients as features.

J C Brown1

  • 1Physics Department, Wellesley College, Massachusetts 02181, USA. brown@media.mit.edu

The Journal of the Acoustical Society of America
|March 25, 1999
PubMed
Summary

This study used cepstral coefficients and pattern analysis to accurately classify short oboe and saxophone sounds. Machine learning algorithms achieved high accuracy, comparable to human perception.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Prior self-control exertion decreases pre-frontal cortex oxygenation during a CO<sub>2</sub> rebreathing challenge but does not affect perceptions of dyspnoea or tolerance time.

Respiratory physiology & neurobiology·2024
Same author

A Case of Corrosive Poisoning: In Which the Mucous Membrane of Esophagus and a Portion of the Stomach Was Vomited as a Complete Cast.

Buffalo medical journal·2023
Same author

Presenting features of female collegiate sports-related concussion in South Africa: a descriptive analysis.

South African journal of sports medicine·2023
Same author

Polymer formulated self-amplifying RNA vaccine is partially protective against influenza virus infection in ferrets.

Oxford open immunology·2022
Same author

The effect of trait self-control on dyspnoea and tolerance to a CO<sub>2</sub> rebreathing challenge in healthy males and females.

Physiology & behavior·2022
Same author

Body Size Influences Stingless Bee (Hymenoptera: Apidae) Communities Across a Range of Deforestation Levels in Rondônia, Brazil.

Journal of insect science (Online)·2019

Area of Science:

  • Acoustics
  • Music Information Retrieval
  • Machine Learning

Background:

  • Distinguishing between musical instrument sounds is crucial for music information retrieval and analysis.
  • Automated classification of instrument sounds can aid in musicological studies and digital archiving.

Purpose of the Study:

  • To develop and evaluate a machine learning system for classifying short audio segments of oboe and saxophone sounds.
  • To compare the performance of the automated system against human listeners in instrument identification.

Main Methods:

  • Cepstral coefficients were extracted from short oboe and saxophone sound samples using a constant Q transform.
  • A k-means clustering algorithm was applied to longer training sounds to model instrument characteristics.
  • Gaussian probability density functions and a Bayes decision rule were employed for sound classification.

Related Experiment Videos

Main Results:

  • The pattern analysis system demonstrated high accuracy in classifying short oboe and saxophone sounds.
  • Classification performance was comparable to results obtained from a human perception experiment on a subset of the sounds.

Conclusions:

  • Cepstral coefficients derived from constant Q transform are effective features for automated oboe and saxophone sound classification.
  • The developed machine learning approach provides a reliable method for instrument identification, rivaling human capabilities.