Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Sep 27;11(1):19149.
doi: 10.1038/s41598-021-98742-x.

Identifying individuals with recent COVID-19 through voice classification using deep learning

Pichatorn Suppakitjanusant  1 , Somnuek Sungkanuparph  1 , Thananya Wongsinin  1 , Sirapong Virapongsiri  1 , Nittaya Kasemkosin  2 , Laor Chailurkit  3 , Boonsong Ongphiphadhanakul  4
Affiliations

Identifying individuals with recent COVID-19 through voice classification using deep learning

Pichatorn Suppakitjanusant et al. Sci Rep. .

Abstract

Recently deep learning has attained a breakthrough in model accuracy for the classification of images due mainly to convolutional neural networks. In the present study, we attempted to investigate the presence of subclinical voice feature alteration in COVID-19 patients after the recent resolution of disease using deep learning. The study was a prospective study of 76 post COVID-19 patients and 40 healthy individuals. The diagnoses of post COVID-19 patients were based on more than the eighth week after onset of symptoms. Voice samples of an 'ah' sound, coughing sound and a polysyllabic sentence were collected and preprocessed to log-mel spectrogram. Transfer learning using the VGG19 pre-trained convolutional neural network was performed with all voice samples. The performance of the model using the polysyllabic sentence yielded the highest classification performance of all models. The coughing sound produced the lowest classification performance while the ability of the monosyllabic 'ah' sound to predict the recent COVID-19 fell between the other two vocalizations. The model using the polysyllabic sentence achieved 85% accuracy, 89% sensitivity, and 77% specificity. In conclusion, deep learning is able to detect the subtle change in voice features of COVID-19 patients after recent resolution of the disease.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
Mel-spectrogram of the 3 voice types from a study subject.
Figure 2
Figure 2
Shannon entropy of the 3 voice types.
See this image and copyright information in PMC

References

    1. Li MY, Li L, Zhang Y, Wang XS. Expression of the SARS-CoV-2 cell receptor gene ACE2 in a wide variety of human tissues. Infect. Dis. Poverty. 2020;9:45. doi: 10.1186/s40249-020-00662-x. - DOI - PMC - PubMed
    1. Jia HP, et al. ACE2 receptor expression and severe acute respiratory syndrome coronavirus infection depend on differentiation of human airway epithelia. J. Virol. 2005;79:14614–14621. doi: 10.1128/jvi.79.23.14614-14621.2005. - DOI - PMC - PubMed
    1. Xu H, et al. High expression of ACE2 receptor of 2019-nCoV on the epithelial cells of oral mucosa. Int. J. Oral Sci. 2020;12:8. doi: 10.1038/s41368-020-0074-x. - DOI - PMC - PubMed
    1. Lechien JR, et al. Features of mild-to-moderate COVID-19 patients with dysphonia. J. Voice Off. J. Voice Found. 2020 doi: 10.1016/j.jvoice.2020.05.012. - DOI - PMC - PubMed
    1. Lin KWE, Balamurali BT, Koh E, Lui S, Herremans D. Singing voice separation using a deep convolutional neural network trained by ideal binary mask and cross entropy. Neural Comput. Appl. 2020;32:1037–1050. doi: 10.1007/s00521-018-3933-z. - DOI