COVID-19 Patient Detection from Telephone Quality Speech Data
This addresses a potential non-invasive screening method for COVID-19 detection, but it is incremental as it applies existing methods to a new dataset.
The paper tackled detecting COVID-19 patients from telephone-quality speech by using a speaker recognition-like approach with super vectors of Mel filter bank features, achieving an accuracy of 88.6% and an F1-score of 92.7% on a small YouTube dataset.
In this paper, we try to investigate the presence of cues about the COVID-19 disease in the speech data. We use an approach that is similar to speaker recognition. Each sentence is represented as super vectors of short term Mel filter bank features for each phoneme. These features are used to learn a two-class classifier to separate the COVID-19 speech from normal. Experiments on a small dataset collected from YouTube videos show that an SVM classifier on this dataset is able to achieve an accuracy of 88.6% and an F1-Score of 92.7%. Further investigation reveals that some phone classes, such as nasals, stops, and mid vowels can distinguish the two classes better than the others.