A Meter Classification System for Spoken Persian Poetries

In this article, a meter classification system has been proposed for Persian poems based onfeatures that are extracted from uttered poem. In the first stage, the utterance has beensegmented into syllables using three features, pitch frequency and modified energy of each frameof the utterance and its temporal variations. In the second stage, each syllable is classified intolong syllable and short syllable classes which is a historically convenient categorization in Persianliterature. In this stage, the classifier is an SVM classifier with radial basis function kernel. Theemployed features are the syllable temporal duration, zero crossing rate and PARCORcoefficients of each syllable. The sequence of extracted syllables classes is then softly comparedwith classic Persian meter styles using dynamic time warping, to make the system robust againstsyllables insertion, deletion or classification and poems authorities. The system has beenevaluated on 136 poetries utterances from 12 Persian meter styles gathered from 8 speakers,using k-fold evaluation strategy. The results show 91% accuracy in three top meter style choicesof the system.