Recognition of Imitative Tehrani Speech from a Standard Persian through Phonological Analysis of the Intonation Pattern in the Framework of the Taylor Tilt Model

The present study intends to investigate a set of acoustic parameters extracted from the intonation pattern of Kashani and Tehrani accents based on the acoustic approach and in the framework of Accent Forensic Compariton using the Tyler the tilt model, in order to Introduce the most appropriate acoustic parameters that differentiate Imitative Tehrani (a speech in which Kashani's speaker tries to speak as close to Tehrani as possible) from standard Persian. For this purpose, 84 5-6 minute two-person conversations using ZOOM H5 professional voice recorder and Praat software from 28 speakers (14 men and women from Kashan + 14 men and women from Tehran) was recorded in space as quiet as possible. Then, 756 utterances were extracted, and their text grid was manually layered and labeled to measure and quantify acoustic correlations within the Taylor tilt model. Objective and statistical results showed that the introduced acoustic parametres have the potential to distinguish between the original Tehrani and Imitative Tehrani in the question utterances, but the tilt values for affirmative utterances is not differentiating.