Effects of Data Augmentation on the Identification of Cough Sound Using Convolutional Neural Networks

David Naranjo, Juan Chica, Christian Salamea Palacios

Resultado de la investigación: Capítulo del libro/informe/acta de congresoContribución de conferenciarevisión exhaustiva

Resumen

Cough identification employing deep learning models has generated significant interest in fields such as telemedicine. However, it is challenging to have a considerable number of cough sounds to train such models, and consequently, it is necessary to apply techniques to generate more data. In this work, the effect of data augmentation on the identification of cough was studied using convolutional neural networks (CNN). For this, a system composed of Front-End and Back-End was implemented. In the Front-End, the original recordings and those generated by data augmentation by frequency shift and noise injection were processed and transformed into Mel spectrograms. On the Back-End, the spectrograms were analyzed and classified using a CNN. The results show that the transformations proposed to increase the amount of data help reducing the overfitting by 42.5% while bettering accuracy by a relative improvement of 1%. Therefore, applying data augmentation techniques to train CNNs for differentiating coughs from other sound events results in a simple but effective strategy.

Idioma originalInglés
Título de la publicación alojadaCommunication, Smart Technologies and Innovation for Society - Proceedings of CITIS 2021
EditoresÁlvaro Rocha, Paulo Carlos López-López, Juan Pablo Salgado-Guerrero
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas285-295
Número de páginas11
ISBN (versión impresa)9789811641251
DOI
EstadoPublicada - 2022
Evento7th International Conference on Science, Technology and Innovation for Society, CITIS 2021 - Virtual, Online
Duración: 26 may. 202128 may. 2021

Serie de la publicación

NombreSmart Innovation, Systems and Technologies
Volumen252
ISSN (versión impresa)2190-3018
ISSN (versión digital)2190-3026

Conferencia

Conferencia7th International Conference on Science, Technology and Innovation for Society, CITIS 2021
CiudadVirtual, Online
Período26/05/2128/05/21

Nota bibliográfica

Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Huella

Profundice en los temas de investigación de 'Effects of Data Augmentation on the Identification of Cough Sound Using Convolutional Neural Networks'. En conjunto forman una huella única.

Citar esto