Acoustic Characteristics of Pakistani English Vowel Sounds

The purpose of this investigation was to document acoustic characteristics of Pakistani English (PaKE) vowel sounds. The experiment was designed to examine the properties of ten vowels produced by Pakistani ESL learners. The analysis is based on the voice samples of recorded 50 CVC words. Total 5000 (10× 10 × 50=5000) voiced samples were analyzed. The data consisted of 50 words of ten English vowel sounds [i: ɪ e ɔ: æ ə ɑ: u: ɒ ʊ]. Ten ESL speakers recorded their voice samples on Praat speech processing tool installed on laptop. Three parameters were considered i.e., fundamental frequency (F0), vowel quality (F1-F2) and duration. Formant patterns were judged manually by visual inspection on Praat Speech Processing Tool. Analysis of formant frequency shows numerous differences between male and female of F1 and F2, fundamental frequency and duration of English vowels. The voice samples provide evidence for higher and lower frequency of vowel sounds. Additionally, the data analysis illustrates that there were statistical differences in the values of short and long vowels coupled with vowel space plot showing explicit differences in locating the production of vowels of male & female vowel space acoustic realizations.


Introduction
This study is based on 5000 voice samples (10× 10 × 50=5000) which were analyzed in a very careful manner.The experiment was designed to document English vowel properties produced by Pakistani English speakers.To analyze PaKE vowels, three phonetic exponents were considered as parameters for the analysis e.g., duration, fundamental frequency and first and second formant frequency.The study focuses on the characterization of vowel sounds produced by English as Second Language (ESL) learners and the positions of Pakistani vowels space in vowel space plot of mouth cavity.The objectives behind this study are to document the variations in frequencies of 10 different English vowels by ESL learners.Every vowel has different frequency and the vowel space plot when it is pronounced by different people it contains a different frequency, plot, and duration coupled with different pitch.Praat speech processing tool was utilized to analyze the sounds with the help of spectrographic view.Praat calculated the average of all 5000 sounds.The measurements were categorized in terms of male, female and third their measurements were compared with each other.The study tests the hypothesis that if ESL learners pronounce English vowels, then their placement of vowels on vowel space will be differently placed by male and female Pakistani ESL learners.Based on the research questions as follows: What are the acoustic realizations of PaKE English vowel sounds?What is difference in placement of acoustic vowel space plot between male and female?
Additionally, the study focuses on acoustic properties of English vowel sounds and their acoustic vowel space in vowel inventory plot.The vowels are basically categorized referencing three phonetic factors i.e., height of the vowel, back-ness and lip rounding.Olive, Greenwood, & Coleman (1993) contends "An acoustic analysis of vowels stresses the different formant configurations that are characteristic of each vowel.The relationship among the vowels can be examined by comparing their formant values".Front-back and high-low differences are characterized by the second and the first formants, illustrating on spectrographic images.If the formant value is high, the closer the vowel becomes to the front point.In so far as front vowel is concerned frequency F1 turns out to be lower and contraction escalates in the vocal track.F1 and F2 frequencies are regarded as a vowel quality in acoustic speech terms.Generally unstressed vowels are reduced in quality; whereas the stressed vowels have higher frequencies.There are several features in the production of sounds, i.e., reduction of vowel quality, vowel lengthening followed by the voiced sound.Another important factor is absence and presence of lexical stress which affects the vowel quality.The unstressed syllables in American English are reduced in quality therefore, they involve in vowel centralization to schwa vowel sound, the chiefly produced sound in English language.F1 and F2 formants have greater distance when there occurs in between the highest vocal sounds.Whereas, for back vowel, F2 is much closer and lower to F1 for the back vowels than for the front vowels (Parson, 1987).

Procedures of Data Collection
The data was collected from 10 subjects (5 male and 5 female).Their age ranged from 17 to 22 years.The recordings were made in a silent zone where there was no background noise.ESL learners were given 50 indexing cards written on them the token phrase in between the token words, all ten vowels were placed as illustrating Table 1.Ten vowels were selected as follows: /i:, ɑ:, ɪ, ɔ:, ae, ə, u:, ɒ, e, ʊ/.Then five words from each vowel sound and then written them on the cards.Then we have given those cards to our respected subjects and then well requested them to pronounce each word three times.Each subject pronounces the 50 words three times like I say token word.

Speech Material
Five words from each vowel sound were chosen for the experiment.Total chosen words were mono-syllable words.The words frequency in daily use was highly familiar to ESL learners.Table 1 illustrates the fifty mono-syllable words which were used as the data tool as follows:

Recordings
Recordings were made of /hVd/ utterances recorded by ten male and female speakers.Measurements were analyzed of vowel duration, F0 contours (fundamental frequency), and formant frequency contours (F1-F2).A set of 50 cards with each word written on them was provided among the participating subjects.Total fifty words (25 for short & 25 for long vowels) were involved.The word structure was selected where each vowel sound was amid the word.There was a total of 50 cards, i.e., one card for each word inscribing "I say token word now" in a phrase.In addition, the set of 50 cards was read with 03 repetitions by each ESL learners.The cards were reshuffled after each recording, and the subjects read aloud the word on each card in sequence with no break between repetitions.In addition, the participants were provided a quiet and calm environment, with no background noise, in a closed room far from the noise of city traffic.The voice recorder used faced them at a distance of around six inches.They were instructed to maintain a normal speech rate and to read the given words using a neutral focus, as in ordinary talking conditions.Subjects voiced were recorded on sound recorder and analyzed on computer Dell Core 2 dual system using the Praat speech processing tool (Boersma & Weenink, 2017) and a voice recorder. ijel.ccsenet.

Data A
The recor parameters taken as m were on pu the data f carefully.duration a calculated belongs to got the plo

Measu
All measu examinatio were meas using the frequency mid-point in the spe display on Measurem millisecon as below.

Overal
The result research as of the male of the fem average F0 the male sp pitch and d shows that the smalle value (190 male and f English vo illustrated correspond frequency within the Female pro and in hei vowel spac ʊ/ are loca the back.

Conclus
The study variety spo view of s parameters male subje subject ha (170 Hz) a compared PaKE prod space acou

Table 1 .
Token words produced by ESL learners