Identification of Vowel Phonemes for Speech Correction Using PRAAT Scripting and SPPAS

Today's advancement in computational linguistics is quite astounding especially in the realm of speech recognition. However, Computer aided language learning (CALL) field of study is still considered a challenge for computational linguistics. Vowels form the basic pronunciation structure of a word and are most important in performing pronunciation correction and evaluation. This paper attempts to identify vowel phonemes in a given word based on machine learning algorithms. Identified vowels can be later analyzed and evaluated in order to provide pronunciation correction for native people learning non-native languages.