**1. Introduction**

The purpose of this paper is to develop the frequency domain of the study started in [1]. In particular, we present an algorithm which obtains the first two formants (*F*1 and *F*2) of a vowel segment. These two elements are most often enough to disambiguate an English vowel, being crucial for non-native speakers' oral training. *F*1 and *F*2, corresponding to mouth opening and tongue position respectively, provide the necessary information for a proficient pronunciation. The phonological information rendered by *F*1 and *F*2 frequency contents produces an algorithm which can help non-native students of English in positioning their tongue and lips.
