Volume 10, Issue 4 (3-2024)
Human Information Interaction 2024, 10(4): 56-75 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Emamverdi* M, Fahimnia F, FahimiFar S. Investigating the improvement of reading experience and phonetic transcriptions of non-Persian speakers: the case study of Saadi's poems. Human Information Interaction 2024; 10 (4)
URL: http://hii.khu.ac.ir/article-1-3145-en.html
URL: http://hii.khu.ac.ir/article-1-3145-en.html
Associate Professor; Department of Knowledge and Information Science; University of Tehran; Tehran; Iran;
Abstract: (2783 Views)
Background and Purpose: In digital humanities, citizens encounter in the creation and consumption of cultural resources, with activities such as transcription and transliteration of text, text correction, punctuation, Classification, description of metadata, labeling, division and annotation of features leading to the enrichment of digitization or production Digital corpuses with different types of annotations. Digital Annotation Tools (DAT) facilitate the development of new digital-based reading strategies and allow learners to highlight digital texts and add textual, video, and image annotations. Social annotation can become a "novice art" for teaching literature. Also, citizen science projects emphasize that education should be considered as a high priority goal. The successful development of a practical method to use the collaborative annotation system to increase the academic output of digital humanities researchers and improve reading in the digital space is still a valuable research topic. Computer-aided transcription tools can speed up the process of reading and transcribing texts. An automatic method for converting Persian texts into phonetic chains has been presented, but in this context, the problem is not limited to automatic phonetic transcription, where the absence of short vowel letters (short vowels) in replacing Arabic letters with Latin characters (letters) leads to poor readability of transliterated text. Reading skill requires many prerequisites, one of them is phonological awareness. The purpose of this research was to Investigate the changes in users' learning and reading experience using annotation types and Investigate the impact of international transliteration training on users' learning and reading experience.
Materials and methods: The approach of this research was quantitative and the research method was semi-experimental and its design was pre-test-post-test. The statistical population includes 30 non-Persian speakers who had no familiarity with the Persian language. In order to reach the target audience and distribute the questionnaires, some Persian language teaching institutions such as Dehkhoda Institute, Persian language teaching center of universities. Also, the sites of some scientific social networks such as Research Gate, Linkedin and Academia were referred and the profiles of people were observed. In order to carry out the research, 30 non-Persian speaking people were selected from the statistical population using a non-probability and targeted sampling method and were randomly assigned into two experimental and control groups (15 people in the experimental group and 15 people in the control group). In terms of gender, 12 were men and 18 were women (40% men and 60% women). In terms of age group, most of them were between 19 and 50 years old, and the average age was 33 years old. In terms of nationality, 2 Iraqis, 2 Americans, 5 Chinese, 2 French, 3 German, 5 Ukrainian, 1 English, 1 Swiss, 3 Japanese, 1 Canadian, 2 Andalusian, 2 Italian and 1 person was Indian. In terms of education level, 12 people were in the doctoral level, 9 people were in the master's level, and 6 people were in the bachelor's level. An interface was designed to collect data. a few verses of Saadi's poems in which words that have the same written appearance but have completely different pronunciations, meanings and grammatical characteristics, were searched and selected. Depending on whether there are different options for pronouncing words or not, we deliberately included possible but incorrect pronunciations for some random words in the question so that they would listen to the voice of the local announcer and then choose one from among the transliterated words of the poem. In each pre-test and post-test stage, users were faced with 23 disambiguation items (in the form of disambiguation by clicking on the correct option among multiple choice options) and 2 correction items (by typing the correct form in the corresponding field). For all 23 questions, there were real words in the Persian language.To design the intervention activity of international transliteration education, the Persian book of the first year of elementary school and the second and third sections, which include the signs of the Persian language, were used. The structure of each sign (letter) in the book to read consists of five parts: big picture - small pictures with the word - written text - new sign - learning activities. Under each sign and thumbnails of each word, international transliteration and audio file were added. The training files were provided to the experimental group and trainings were provided in 4 sessions of 30 minutes and Skype and Google Meet communication software were used. Users of both groups were evaluated twice (before and after the intervention of international phonetic training). The tools used in this research were two questionnaires to perform practical work, one of which was used in the pre-test phase and the other in the post-test phase. After the pre-test, the experimental group received an interactive training support introducing the work program and IPA (International Phonetic Association, 1999) in order to improve the learning experience and develop the educational structure, while the control group did not receive this intervention during the research process. To calculate the distance between the answer chosen by the user and the correct answer in the poem, the method of calculating phonetic distances by Faradi et al (2019) was used.
Findings: For analysis, three independent variables were identified for each item (question); Two of them were interval variables, i.e., word length and Lonstein's phonological weighted distance between the selected answer and the correct answer, the other parameter was a nominal parameter that indicated whether the correct answer was among the selected options or not. The average scores in the learning and user reading experience of the control group in the pre-test was equal to 18.6 and in the post-test it was equal to 19.8, while the average scores in the learning and user reading experience of the experimental group before the international transliteration training was equal to 19.1 and after the international transliteration training, it increased to 23.2. For this purpose, the t-test of independent groups was used after the international transliteration training. The result of independent group t-test indicates the change of scores in users' learning and reading experience between the two experimental and control group after the international transliteration training. The resulting t (3.08) is greater than the critical table's t (1.96), so it can be said that there is a significant difference between the scores in the learning and reading experience of the users between the two experimental and control groups after the international transliteration training (p 0.05 < ). Therefore, according to the evidence, it can be said that the types of annotations (audio, transliteration) have resulted in a significant change in the learning and reading experience of users. In order to answer the second question, "What effect does international transliteration training have on users' experience in reading Persian language poems?" In the descriptive analysis of the information, first the statistical indicators related to the basic variables of the research were calculated. The results showed that in the pre-test stage, the average distance from the correct answer (error rate) of the experimental group was 4.09 and in the post-test stage it reached 1.1, which shows a significant decrease. In the control group, the average distance from the correct answer (error rate) was equal to 3.4 and in the post-test stage it reached 3.6, which did not change much. According to the quasi-experimental design of this research and the use of pre-test and post-test in it and in order to control the effect of the pre-test in it, to analyze the relevant data, the statistical model of covariance analysis has been implemented. Before running any covariance analysis, it was necessary to have homogeneous regression slopes and a linear relationship between the auxiliary random variable and the dependent variable. For this purpose, the interaction between the random variable and the independent variable has been investigated. On this basis, the assumptions of this statistical method, i.e. the assumption of normality of distribution, equality of error variances and homogeneity of coefficients and regression, have been investigated. In the inferential methods section, in order to answer the second question, "What effect does international transliteration training have on the experience of users in reading Persian language poems?" Between-group covariance analysis was used, and the results are presented in separate tables. Due to the fact that the studied groups have been selected completely randomly in compliance with all conditions, therefore, the distribution of the data was completely normal. Another important assumption of covariance analysis is the homogeneity of regression coefficients. Based on the results, it was observed that the interaction of the pre-test distance from the correct answer (error rate) with the independent variable is significant at the confidence level of 0.95 (p < 0.05). One of the assumptions of covariance analysis is the assumption of equality of error variances (homogeneity of variances), which Levin's F was used to check this assumption (to check the post-test of the dependent variable). Based on the reported results, it was observed that the homogeneity of variances is significant at the confidence level of 0.95 (p < 0.05). In order to analyze the results and changes of the dependent scale of the research, univariate covariance analysis between groups was used. The results showed that considering the pre-test scores "distance from the correct answer (error rate)", the difference between the trained group and the control group is significant in the post-test stage. (p<0.05). Distance from the correct answer (error rate) (F = 12.3). The total effect size is equal to 0.314. Considering the eta square, it can be said that 31.4% of the changes and the reduction of the distance from the correct answer (error rate) are caused by the influence of the independent variable (i.e. intervention and international phonetic training), which is in accordance with Cohen's criteria, close to It is moderate and statistically significant. (p<0.05). Therefore, it can be said that international phonetic training has been effective on the distance from the correct answer (error rate).
Conclusion: Level-based achievement was confirmed in the results of the users' learning and reading tests. Annotations act as an attention strategy. Vocabulary marking and adding various annotations and international transliteration were effective for non-Persian speakers to learn and read Persian poetry.The results of this research also highlight the important role of international phonetic training in phonetic transcription and improving the second language reading experience and show the effectiveness of international phonetic training intervention and phonological memory training as a tool to strengthen second language vocabulary development. The limitations of this research include ignoring the influence of the users' native languages, the difference in the phonological systems of different languages, relatively small sample size, different age groups and different education levels, the influence and relationships of these factors. The output of the text-to-phonetic chain conversion program is used in automatic text-to-speech conversion, correct phonetic writing of texts, teaching Persian to non-Persian speakers, dictionary writing, etc. By performing phonetic transcription work in the form of manual disambiguation or correction, users create a database with possible corrections that are suggested to be used for various purposes, including increasing the accuracy of machine transliteration. Using a digital annotation tool that allows more accurate tracking of how each user interacts with the text (through log files, tracking eye movements in the number of times focusing on a word, attention creation mechanisms) Important points for future empirical work in this field. Implementation of computer-based assessment using machine learning method leads to efficient implementation and improvement of PA skills. Although solutions for computer-based phonological awareness assessment for children have been provided in Persian language, computer-based phonological awareness assessment can also be used in teaching reading literature. Implementation of computer-based assessment using machine learning method leads to efficient implementation and improvement of PA skills. Although solutions for computer-based phonological awareness assessment for children have been provided in Persian language, computer-based phonological awareness assessment can also be considered in teaching reading literature.
Materials and methods: The approach of this research was quantitative and the research method was semi-experimental and its design was pre-test-post-test. The statistical population includes 30 non-Persian speakers who had no familiarity with the Persian language. In order to reach the target audience and distribute the questionnaires, some Persian language teaching institutions such as Dehkhoda Institute, Persian language teaching center of universities. Also, the sites of some scientific social networks such as Research Gate, Linkedin and Academia were referred and the profiles of people were observed. In order to carry out the research, 30 non-Persian speaking people were selected from the statistical population using a non-probability and targeted sampling method and were randomly assigned into two experimental and control groups (15 people in the experimental group and 15 people in the control group). In terms of gender, 12 were men and 18 were women (40% men and 60% women). In terms of age group, most of them were between 19 and 50 years old, and the average age was 33 years old. In terms of nationality, 2 Iraqis, 2 Americans, 5 Chinese, 2 French, 3 German, 5 Ukrainian, 1 English, 1 Swiss, 3 Japanese, 1 Canadian, 2 Andalusian, 2 Italian and 1 person was Indian. In terms of education level, 12 people were in the doctoral level, 9 people were in the master's level, and 6 people were in the bachelor's level. An interface was designed to collect data. a few verses of Saadi's poems in which words that have the same written appearance but have completely different pronunciations, meanings and grammatical characteristics, were searched and selected. Depending on whether there are different options for pronouncing words or not, we deliberately included possible but incorrect pronunciations for some random words in the question so that they would listen to the voice of the local announcer and then choose one from among the transliterated words of the poem. In each pre-test and post-test stage, users were faced with 23 disambiguation items (in the form of disambiguation by clicking on the correct option among multiple choice options) and 2 correction items (by typing the correct form in the corresponding field). For all 23 questions, there were real words in the Persian language.To design the intervention activity of international transliteration education, the Persian book of the first year of elementary school and the second and third sections, which include the signs of the Persian language, were used. The structure of each sign (letter) in the book to read consists of five parts: big picture - small pictures with the word - written text - new sign - learning activities. Under each sign and thumbnails of each word, international transliteration and audio file were added. The training files were provided to the experimental group and trainings were provided in 4 sessions of 30 minutes and Skype and Google Meet communication software were used. Users of both groups were evaluated twice (before and after the intervention of international phonetic training). The tools used in this research were two questionnaires to perform practical work, one of which was used in the pre-test phase and the other in the post-test phase. After the pre-test, the experimental group received an interactive training support introducing the work program and IPA (International Phonetic Association, 1999) in order to improve the learning experience and develop the educational structure, while the control group did not receive this intervention during the research process. To calculate the distance between the answer chosen by the user and the correct answer in the poem, the method of calculating phonetic distances by Faradi et al (2019) was used.
Findings: For analysis, three independent variables were identified for each item (question); Two of them were interval variables, i.e., word length and Lonstein's phonological weighted distance between the selected answer and the correct answer, the other parameter was a nominal parameter that indicated whether the correct answer was among the selected options or not. The average scores in the learning and user reading experience of the control group in the pre-test was equal to 18.6 and in the post-test it was equal to 19.8, while the average scores in the learning and user reading experience of the experimental group before the international transliteration training was equal to 19.1 and after the international transliteration training, it increased to 23.2. For this purpose, the t-test of independent groups was used after the international transliteration training. The result of independent group t-test indicates the change of scores in users' learning and reading experience between the two experimental and control group after the international transliteration training. The resulting t (3.08) is greater than the critical table's t (1.96), so it can be said that there is a significant difference between the scores in the learning and reading experience of the users between the two experimental and control groups after the international transliteration training (p 0.05 < ). Therefore, according to the evidence, it can be said that the types of annotations (audio, transliteration) have resulted in a significant change in the learning and reading experience of users. In order to answer the second question, "What effect does international transliteration training have on users' experience in reading Persian language poems?" In the descriptive analysis of the information, first the statistical indicators related to the basic variables of the research were calculated. The results showed that in the pre-test stage, the average distance from the correct answer (error rate) of the experimental group was 4.09 and in the post-test stage it reached 1.1, which shows a significant decrease. In the control group, the average distance from the correct answer (error rate) was equal to 3.4 and in the post-test stage it reached 3.6, which did not change much. According to the quasi-experimental design of this research and the use of pre-test and post-test in it and in order to control the effect of the pre-test in it, to analyze the relevant data, the statistical model of covariance analysis has been implemented. Before running any covariance analysis, it was necessary to have homogeneous regression slopes and a linear relationship between the auxiliary random variable and the dependent variable. For this purpose, the interaction between the random variable and the independent variable has been investigated. On this basis, the assumptions of this statistical method, i.e. the assumption of normality of distribution, equality of error variances and homogeneity of coefficients and regression, have been investigated. In the inferential methods section, in order to answer the second question, "What effect does international transliteration training have on the experience of users in reading Persian language poems?" Between-group covariance analysis was used, and the results are presented in separate tables. Due to the fact that the studied groups have been selected completely randomly in compliance with all conditions, therefore, the distribution of the data was completely normal. Another important assumption of covariance analysis is the homogeneity of regression coefficients. Based on the results, it was observed that the interaction of the pre-test distance from the correct answer (error rate) with the independent variable is significant at the confidence level of 0.95 (p < 0.05). One of the assumptions of covariance analysis is the assumption of equality of error variances (homogeneity of variances), which Levin's F was used to check this assumption (to check the post-test of the dependent variable). Based on the reported results, it was observed that the homogeneity of variances is significant at the confidence level of 0.95 (p < 0.05). In order to analyze the results and changes of the dependent scale of the research, univariate covariance analysis between groups was used. The results showed that considering the pre-test scores "distance from the correct answer (error rate)", the difference between the trained group and the control group is significant in the post-test stage. (p<0.05). Distance from the correct answer (error rate) (F = 12.3). The total effect size is equal to 0.314. Considering the eta square, it can be said that 31.4% of the changes and the reduction of the distance from the correct answer (error rate) are caused by the influence of the independent variable (i.e. intervention and international phonetic training), which is in accordance with Cohen's criteria, close to It is moderate and statistically significant. (p<0.05). Therefore, it can be said that international phonetic training has been effective on the distance from the correct answer (error rate).
Conclusion: Level-based achievement was confirmed in the results of the users' learning and reading tests. Annotations act as an attention strategy. Vocabulary marking and adding various annotations and international transliteration were effective for non-Persian speakers to learn and read Persian poetry.The results of this research also highlight the important role of international phonetic training in phonetic transcription and improving the second language reading experience and show the effectiveness of international phonetic training intervention and phonological memory training as a tool to strengthen second language vocabulary development. The limitations of this research include ignoring the influence of the users' native languages, the difference in the phonological systems of different languages, relatively small sample size, different age groups and different education levels, the influence and relationships of these factors. The output of the text-to-phonetic chain conversion program is used in automatic text-to-speech conversion, correct phonetic writing of texts, teaching Persian to non-Persian speakers, dictionary writing, etc. By performing phonetic transcription work in the form of manual disambiguation or correction, users create a database with possible corrections that are suggested to be used for various purposes, including increasing the accuracy of machine transliteration. Using a digital annotation tool that allows more accurate tracking of how each user interacts with the text (through log files, tracking eye movements in the number of times focusing on a word, attention creation mechanisms) Important points for future empirical work in this field. Implementation of computer-based assessment using machine learning method leads to efficient implementation and improvement of PA skills. Although solutions for computer-based phonological awareness assessment for children have been provided in Persian language, computer-based phonological awareness assessment can also be used in teaching reading literature. Implementation of computer-based assessment using machine learning method leads to efficient implementation and improvement of PA skills. Although solutions for computer-based phonological awareness assessment for children have been provided in Persian language, computer-based phonological awareness assessment can also be considered in teaching reading literature.
Keywords: digital annotation, interactive learning environments, user reading experience, phonetic transcription, digital humanities, crowdsourcing, disambiguation, citizen science
Send email to the article author
| Rights and permissions | |
 | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
