An Algorithm for Smoker Detection
Hiba Youni, Saleem Omar, Mafaz Alwahhab, Enas Mqdad · Maǧallaẗ al-rāfidayn li-ʿulūm al-ḥāsibāt wa-al-riyāḍiyyāẗ/Al-Rafidain journal for computer sciences and mathematics · 2013
In this research an algorithm was suggested for studying speech signal properties for both smokers and non smokers then verificate that the person smoker or not based on his speech signal. A data base that contain 30 speech signals 15 belong to smoker and 15 belong to non smokers for male only. In this algorithm formant frequencies such as f1, f2 were adopted as characteristic properties for speech signal for splitting between two classes which it calculate using lpc algorithm. The algorithm consist of two stages: ♦ Data base preparation stage ♦ Speaker state classification stage The absolute, eclideance and d1 distance were adopted as measures for evaluating the performance of the algorithm and it gave convergece results.