Analysis of COVID-19 Resulting Cough Using Formants and Automatic Speech Recognition System.
Ouissam Zealouk, Hassan Satori, Mohamed Hamidi and 3 others
PMID 34256982WHAT IT FOUND
A prototype speech-recognition system reported cough recognition rates of 93.33% and 86.66% in two small experiments.
Pitch and formant values differed, but the study was too small to support a clinical cough test.
Key findings
01The study recorded cough sounds from 10 people: 5 healthy people and 5 people with COVID-19.
02The cough recognition system reported rates of 93.33% and 86.66% in two experiments.
03The authors state that the database size may have influenced the results.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Only 10 people were included: 5 healthy and 5 with COVID-19. The recognition tests used only three speakers from each group. The authors state that the database size may have influenced the results. The authors report difficulty reaching people during symptom onset and recording natural cough from volunteers.
The easy way to misread this
Do not read the 93.33% and 86.66% recognition rates as evidence that cough sounds can diagnose COVID-19. The study used 10 people, with three speakers tested in each group, and the authors say the database size may have influenced the results.