Most people in their lives at least once have faked falling sick to take a day off from work. But be warned, as artificial intelligence reaches new heights, this common excuse might not work in getting you a sick leave at work.

As per reports, researchers have decoded how a caller who is sick or has cold symptoms sounds like and used this knowledge to pass on to AI so it can detect a healthy person from a sick one.

For the study, researchers analysed recordings of around 630 different people including 111 people infected with cold. Researchers at the Sardar Vallabhbhai National Institute of Technology made use of the vocal rhythms present in human speech which can be identified as harmonics that diminish in amplitude as their frequency increases.

A harmonic is a wave with a frequency that is a positive integral multiple of the original frequency. The overtone is produced by vibration of a string, column of air etc in an exact fraction of its length.

An algorithm was developed which identifies the specific frequencies present in people with sore throat. As per a report in The Economist, the study, led by electronic engineer Pankaj Warule, tried to understand just how exactly a “cold voice” differs from a healthy voice.

The key element of the research was the fact that human speech, just like music instruments, does not produce single frequencies of sounds.

To find out whether a cold infection alters how attenuation happens, researchers put the components into machine-learning algorithms to detect and study the differences in amplification to find out who has a cold. During the research, the technology could diagnose a cold in 70 per cent of cases. Researchers broke down each person’s speech into its spectrum of component wavelengths to identify the dominant frequency and the harmonics in each case that was part of the study.

The speed at which the technology is progressing, we can expect several corporations and workplaces to implement this algorithm to stop employees from fake falling sick to avoid work. Do you think it’ll work? Let us know in the comments section below.