Wyniki wyszukiwania - BazTech

Ograniczanie wyników

1 TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk

1 2023

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Breath detection from a microphone using machine learning

Sankowski Tomasz, Sulewski Piotr, Bruska Aleksandra, Walczak Jan

TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk

2023

Vol. 27, No 2

1--12

Breathing is a fundamental physiological process that reflects the health and condition of the body. Patterns, depth, and frequency of respiration are critical indicators of an individual’s overall health, with applications ranging from diagnosing illnesses to monitoring stress levels, physical exertion, and sleep quality.This paper investigates and implements various machine-learning techniques for the real-time detection of breath sounds using audio data captured via a computer microphone. The primary objective is to develop and compare methodologies to identify distinct breathing phases, namely inhalation, exhalation, and the silent intervals between breaths, in order to determine the most accurate, efficient, and practical approach. The study explores three approaches: 1. VGGish Model for Feature Extraction and Classification with Random Forest. 2. Spectrogram Classification Using Convolutional Neural Networks. 3. Mel-Frequency Cepstral Coefficients (MFCC) for Feature Extraction and Neural Network Classification.The experimental results show that methods 1 and 3 achieved an accuracy of 87% in the test data, while method 2 achieved an accuracy of 83%. The dataset comprised approximately 1,000 recordings of inhalations, exhalations, and silences between breaths, collected using four different microphones and recorded by three different individuals. All implementations and training data are available on a public GitHub repository: github.com/tomaszsankowski/Breathing-Classification.