Skip to content

Latest commit

 

History

History
19 lines (13 loc) · 1.03 KB

README.md

File metadata and controls

19 lines (13 loc) · 1.03 KB
Raw

Sound Classification with K-Nearest Neighbors (KNN) Algorithm

Overview

This project explores an approach based on the use of MATLAB and the K-Nearest Neighbors (KNN) algorithm to distinguish between different types of sounds. Specifically, the sounds of sneezing, snoring, and crying are analyzed using spectral and temporal features.

Features in Time Domain

  • Energy: The magnitude of the signal.
  • Zero Crossing Rate: The frequency at which the signal changes from positive to negative or back.

Features in Frequency Domain

  • Spectral Centroid: The main point of the spectrum distribution.
  • Spectral Spread: The standard deviation of the spectrum distribution.
  • Spectral Rolloff: The amount of energy accumulated until a certain point in the frequency.
  • Mel-Frequency Cepstral Coefficients (MFCC): Represents the spectrum bands according to the mel-scale, an isophonic (mostly subjective) coefficient.

Credits

This project was developed by Alessandro Scalambrino