Indian Languages Research Articles

Real time Indian sign language recognition using transfer learning with VGG16

Real time Indian sign language recognition using transfer learning with VGG16

Language Policy under NEP: 2020 – A Critical Analysis

Education is the backbone of a country. Education also plays vital role to upload national unity. Thus education policy always plays a congruent role in the spread of education in a country. Education policy acts as a basic requirement of a country. Any education policy sets on the analysis of the past. It deals with the need of the present days and also focuses on the possibilities of the future. Like the other countries, India also felt the need of a well- planned education policy to ensure all-round development of our society at large. The new education policy of India enlightens preservation as well as development of all Indian languages but the problem lays in the fact that India being a multilingual and multicultural country will select which particular language as a medium of instruction in the education system. NEP 2020 for the first time harps on introducing mother tongue as a medium of instruction at the primary level. All the relevant aspects of language issues are carefully integrated from paragraphs 4.11 to 4.22 very carefully under the heading ‘Multilingualism’ in NEP 2020. The inclusion of many vernacular languages in educational curriculum will definitely protect endangered languages in India. It will also help the students to be connected with the culture of the nation. Indirectly, this will help our students to increase qualitative creative skills. NEP 2020 has proposed to implement ‘Three-Language-Formula’ to promote multilingualism from the school level. The objective of NEP 2020 is to make the students proficient in different languages and also know the mother tongue to be respectful and knowledgeable about the ancient literature. The objective of this study is to identify the necessity of new education policy (NEP 2020) and its impact on education system of our country for holistic development of the students from primary to higher education. Key Words: Education system, Multilingualism, Multicultural, Holistic Development, Three-Language Formula.

Double-finite Serial Verbs in Sanskrit and later Indo-Aryan

Abstract Steever (1988. The serial verb formation in the Dravidian languages. Delhi: Motilal Banarsidass) introduced to South Asian linguistics the notion “Serial Verb” – a Dravidian construction in which two morphologically finite verbs agree in person and number features in the same sentence. I show that similar “double-finite” structures are found in Sanskrit/Indo-Aryan, although details differ. One double-finite construction, inherited from Proto-Indo-European, does not seem to have continued in the post-Vedic language, but others do, and similar structures occur in Modern Indo-Aryan languages. While the historical developments are relatively clear for early Indo-Aryan, from Vedic down to Pāli, there are questions about whether the double-finite constructions of Modern Indo-Aryan are inherited , since comparable structures have not been noted in late Middle Indo-Aryan Apabhraṁśa, the near-ancestor of (peninsular) Modern Indo-Aryan. In the conclusion I discuss alternatives to Modern Indo-Aryan inheritance, including areal and typological factors and the proposal by Milizia (2018. Sul presente neoindoarico a doppio verbo finito. Atti del Sodalizio Glottologico Milanese 12(n.s.). 93–111) that the perhaps most widespread Modern Indo-Aryan double-finite construction results from analogical developments. (Unfortunately, the latter proposal does not account for the other Modern Indo-Aryan double-finite constructions.)

EXPRESS: Cross-Linguistic Effects of the Speech-to-Song Illusion in Speakers of Bangla and English.

The speech-to-song illusion is a phenomenon in which the continuous repetition of a spoken utterance induces the listeners to perceive it as more song-like. Thus far, this perceptual transformation has been observed in mostly European languages, such as English; however, it is unclear whether the illusion is experienced by speakers of Bangla (Bengali), an Indo-Aryan language. The current study, therefore, investigates the illusion in 28 Bangla- and 31 English-speaking participants. The experiment consisted of a listening task in which participants were asked to rate their perception of repeating short speech stimuli on a scale from 1-5, where 1 = "sounds like speech" and 5 = "sounds like song". The stimuli were composed of English and Bangla utterances produced by two bilingual speakers. To account for possible group differences in music engagement, participants self-reported musical experience and also performed a rhythm discrimination task as an objective measure of non-verbal auditory sequence processing. Stimulus ratings were analysed with cumulative link mixed modelling. Overall, English- and Bangla-speaking participants rated the stimuli similarly and, in both groups, better performance in the rhythm discrimination task significantly predicted more song-like ratings beyond self-reported musical experience. An exploratory acoustic analysis revealed a role of harmonic ratio in the illusion for both language groups. These results demonstrate that the speech-to-song-illusion occurs for Bangla speakers to a similar extent as English speakers and that, across both groups, sensitivity to non-verbal auditory structure is positively correlated with susceptibility to this perceptual transformation.

BPE and morphologically segmented phrase based statistical machine translation system for Indian languages to resource constrained language Bodo

BPE and morphologically segmented phrase based statistical machine translation system for Indian languages to resource constrained language Bodo

Language Preservation through Popularization of Regional Indian Music: A Data-Driven Approach

The increased global connectivity due to internet penetration has resulted in the dominance of international languages and the decline of regional languages. To help preserve these languages, this study identifies the non-native regions in India that offer the most promising opportunities for expanding the popularity of music for each of the 14 regional languages. It analyses the audio features of 41,000 songs from 14 Indian languages on Spotify, employing two approaches – clustering algorithms and Random Forest Regressors. An overlap between the regions identified by both methods for any given language is considered a robust result, providing greater confidence in the results. The results show that while geographical proximity is a significant factor in determining non-native market fits for the music of a given language, alignment in preferences of audio features driven by cultural similarities plays an integral role as well. The findings have important implications for music industry stakeholders, including artists, management, digital platforms, and governments, who can leverage this data to devise strategies to expand the reach of regional music. This includes designing targeted marketing strategies, fostering cross-regional artist collaborations, and optimizing content recommendation algorithms, among others.

Arabic Vocabulary And Its Cultural Influences On The Colloquial Tamil Language: Special Reference To South Tamil Nadu India/ المفردات العربية وتأثيراتها الثقافية في اللغة التاميلية العامية : جنوب ولاية تاميل نادو في الهند أنموذجا

Language is vital for exchanging opinions and ideas, expressing feelings, and fulfilling daily needs. Arabic stands out among other languages due to its various advantages and features. Tamil Nadu, one of the states in India, is known as the historical frame of the Dravidian civilization, one of India's oldest. It is situated in the southernmost part of Asia. Tamil Nadu's residents speak Tamil, an ancient and expressive language celebrated for its rich literature and esteemed position among Indian languages. Tamil Nadu's global prominence stems from its historical oceanic and commercial connections. Centuries ago, the state actively traded with China, Babylonia, Rome, the Levant, the Turks, and, notably, the Arabs. These trade interactions facilitated vigorous import-export activities and improved cultural exchange and integration among diverse peoples. This study focuses on the influence of the Arabic language on Tamil, particularly in contemporary usage, where many Arabic words have been integrated into Tamil vocabulary. This research emphasizes the everyday use of Arabic terms, especially in fields such as law, exemplified by terms like amal, asal, bark, taraf, waterish, wakeel, and wayeetha, etc. It examines the linguistic and cultural similarities and shared usage of words between these two distinct languages. The methodology adopted is qualitative. The primary data was collected through personal interviews and original texts. Secondary data were obtained from related books, and scholarly works and data were analyzed. This study concluded that Arabic influence in colloquial Tamil, especially in southern Tamil Nadu, needs more extensive research. The persistent Arabic influence on Tamil civilization and cultural practices is frequently mistaken for indigenous Tamil elements in academic and administrative contexts. Arabic words used in Tamil have their roots and origins, emphasizing their originality in the Arabic language rather than being borrowed from Tamil.

What can the taxonomy of predicative possession in Malwai Punjabi tell us?

Abstract The study aims to examine the syntactic and semantic behaviors of predicative possession (i.e., have-possessive constructions) in Malwai Punjabi, an underdocumented dialect within the Indo-Aryan language family. Data were collected from longitudinal online interviews with native speakers as consultants, with audio recordings for transcribed target sentences. The results revealed that all the alienable possession, either permanent/temporary or abstract/concrete, could be marked by the postposition koḷ ‘near/with’, whereas inalienable possession, such as whole-part relation and kinship, could not be encoded using koḷ. The prototypicality model and schema-based metaphors explained why koḷ was widely used to express alienable possession in Malwai Punjabi. The analysis of companion and proximity schemata also justified the extended semantics of predicative possession, suggesting a metaphorical mapping of accompaniment and location onto possession. From a typological angle, the case study can not only provide further evidence for the existence of split possession but also contribute to a cognitive understanding of predicative possession in relation to other languages.

7056 Can ChatGPT Educate Multicultural Patients About Metabolic Syndrome?

Abstract Disclosure: G. Wu: None. E. Cheng: None. V.V. Toram: None. K. Rosen: None. A. Wong: None. W. Zhao: None. M. Del Buono: None. Background: Metabolic Syndrome, which affects Asians, is a major risk factor for diabetes. According to the National Institute of Health (NIH), 1 in 3 Americans have Metabolic Syndrome. Diabetes affects 30 million Americans, and 8 million more are undiagnosed. ChatGPT and Bard are Artificial Intelligence (AI) mediated LLMs launched on 11/30/2022 and 3/21/2023, respectively, that generate real-time conversational responses. Purpose: Can ChatGPT educate our multicultural patients about metabolic syndrome? Methods: 1. User prompt into ChatGPT and Bard: What is metabolic syndrome? I am a South Asian male, 40 years old, with a cholesterol level of 250 and a fasting blood sugar level of 120. The exact prompt was used with an Asian male as a substitute. 2. The prompt was translated into Telugu, Hindi, Simplified Chinese, and Traditional Chinese via Google Translate. 3. Flesch-Kincade (FK) method of ascertaining reading level was used. FK is used by the US Department of Education. Results: Flesch-Kincaid reading ease: ChatGPT: English=29.2, Telugu=60.1, Hindi=33.4, Chinese (Trad.)=14.5, Chinese (Simp.)=22.6 Bard: English=35.3, Telugu=52, Hindi=51.7, Chinese (Trad.)=52, Chinese (Simp.)=36.8 Language Group Comparison to English: ChatGPT: Indian Languages p=0.004; Chinese Languages p=0.081. Bard: Indian Languages p=0.007; Chinese Languages p=0.136. Of note, the Telegu response has one incorrect response. Conclusions: In conclusion, ChatGPT and Bard can provide factual responses based on English prompts. Foreign prompts can yield inaccurate information. Physicians and their teams still need to supervise patient usage. Presentation: 6/2/2024

The Qur’an in Malayalam: Contextualising Early iṣlāḥī Translations

This article is an exploration of the texts and historical contexts of the early translations of the Qur'an into the South Indian language of Malayalam. Attempts at Malayalam translations of the Qur'an started in the 1930s. The present study deals with the translation projects that emerged between the 1930s and 1965. It showcases the situatedness of these translations in the colonial experiences of the region, as well as in the discursiveness of global Islamic trends. It also looks at the theological controversies these translations evoked within the Muslim community.

Predictors of Colonoscopy Use Among Asian Indians in New York City, 2003 to 2016.

Asian Americans have the lowest colorectal cancer screening uptake of any racial and ethnic group in the United States. Asian Indians are among the most under-screened Asian American subgroups, but there is limited data for this population. We sought to characterize predictors of colonoscopy use among Asian Indians in New York City. Using 2003 to 2016 data from the New York City Community Health Survey, we identified all Asian Indian participants aged 50 years or older. We examined the association between sociodemographic and medical factors and up-to-date colonoscopy use (defined as colonoscopy within the last 10y) using logistic regression over 4 time periods: 2003 to 2008, 2009 to 2012, 2013 to 2014, 2015 to 2016. On multivariable analysis, language, age, income, recent exercise, body mass index, and influenza vaccination were associated with colonoscopy uptake in 1 time period. Compared with participants who preferred English, those who preferred an Indian language were less likely to have been up-to-date in 2013 to 2014 (odds ratio 0.12, 95% CI 0.02-0.66). Individuals older than 65 years were more likely than those aged 50 to 64 years to have received a colonoscopy in 2009 to 2012 (odds ratio 3.91, 95% CI 1.49-10.24), although the risk estimates were also consistently positive in the other 3 time periods. Among Asian Indians living in New York City, several demographic, socioeconomic, and health-related characteristics predict colonoscopy use. These findings highlight the importance of examining determinants of colonoscopy uptake in this understudied population to inform future public health interventions.

HindiSumm: A Hindi Abstractive Summarization benchmark dataset

Abstractive Text Summarization (ATS) is a task to create a novel summary by generating fresh sentences incorporating new words or rephrasing the article. It is a complex task as the model needs to understand the semantic similarity between the sentences of the text. To fulfill this, there is a need for a large annotated benchmark dataset, which is available for resource-rich languages such as English and non-indic languages. In contrast, for the less-resourced languages, such as Indic languages, the available datasets are limited and involve very short summary sentences. Hence, a language-specific abstractive summarization dataset called HindiSumm was introduced for Hindi, consisting of 570,000 text-summary pairs from Navbharat Times across 21 domains. The HindiSumm dataset’s efficiency is evaluated extrinsically and intrinsically by using various metrics. Furthermore, two recent multilingual-cased pre-trained models are fine-tuned on the HindiSumm dataset individually. In addition, an ensembled approach using weighted averaging is also incorporated to check the efficacy of the proposed dataset. The model is tested with the in-house created dataset, and results are evaluated on ROUGE scores and show significant improvements of around 13.2% for the proposed HindiSumm compared to other benchmark datasets. In the future, the HindiSumm dataset will promote the progress of ATS for the Indian language.

A survey on NLP tasks, resources and techniques for low-resource Telugu-English code-mixed text

With the proliferation of informal content on various social media platforms in the form of posts, comments, and feedback, the importance of analyzing text in code-mixed form is gaining importance. Telugu, a low-resource Indian language, has a lot of online content being generated in code-mixed form. However, the lack of large corpora, annotated data and Natural Language Processing (NLP) resources are impeding research on Telugu-English code-mixed data. This paper provides a survey of existing literature on Telugu-English code-mixed text in the areas of resources, POS tagging, Named Entity Recognition, language identification, sentiment analysis, application tasks, dialog systems, and Question-Answering. Various datasets being used by the researchers in the field, along with methods applied to them are detailed. Research gaps are identified to provide future direction for researchers working in this field.

Writing Confucian History in an Indian Language

Abstract This article uses a case study of an Urdu translation of the classical Chinese Shujing (Book of Documents) to bring to light a process of multiperson, mediated, and ultimately transimperial translation in which European source texts played crucial roles in transmitting detailed knowledge of China to the Indian public sphere. After outlining the theoretical approach of transimperial history and its value for Asian intellectual history, the article positions the Urdu Shujing in relation to earlier Indian and Middle Eastern accounts of China whose triangulated transmission—from East Asia to Europe and thence to South or West Asia—set various precedents for the translation of the Shujing from a prior English translation. The article then focuses on the Urdu version to reveal the transimperial mechanisms it used to not only translate and transliterate classical Chinese, but also to “transcalculate” dates and render Chinese history comparable to both Europe's and India's past. The conclusions argue that, from Ming China via the early modern Catholic empires to the occult subcurrents of British colonial culture, the transimperial pathways along which Confucian history reached India were ultimately Eurasian rather than narrowly inter-Asian in scope. This reliance on earlier European translations would later recur with the spread of “world literature” in the postcolonial era.

Hybrid Convolution-Based Efficientnet-Based Hand Gesture Recognition Framework with Optimized Algorithm

The difficulties in communication and hearing are an important concern for deaf–dumb people, which stop access to their essential and basic needs. Many findings have been made to address sign languages even though this challenging problem is not still solved. Many methods aimed to propose vision-based classifiers through identical pattern investigation tasks by obtaining the difficult handcraft feature descriptions of gestures from the gathered images. However, the efficacy of all those models is less for performing with a huge signbook captured from uncontrolled and complex background conditions. So, an effective Indian Sign Language (ISL) classification method is developed by an advanced deep learning approach. At first, the hand gesture images are obtained from the data source. Only the image of the hand, even from a complicated background, is extracted from the obtained image. The features are extracted using the Scale-Invariant Feature Transform (SIFT) method and Multiscale Vision Transformer (MVT). Then, the extracted features are fed to the Hybrid Convolution-based EfficientNet (HCEN) model. The hyper-parameters in the developed HCEN model are tuned using the implemented Adaptive Political Optimizer (APO) algorithm. The recognized hand signs are obtained from the suggested HCEN model. Various experiments are conducted to determine the performance of the suggested deep learning-based hand gesture recognition model.

A systematic compilation of rating scales developed, translated, and adapted in India.

There is a lack of comprehensive data sources on various rating instruments that have been developed, translated, and adapted in Indian context. Accordinly, this review aimed to compile the available scales/questionnaires/instruments developed, adapted, and translated for use in India. For this, the search engines like PUBMED, Google Scholar, MedKnow, and Science Direct were searched for scales that have been developed, translated, and adapted in Indian context or an Indian language. Only articles reporting a scale/questionnaire development/interview schedules from India or in an Indian language were included. Available data suggests that most of the instruments that have been translated in the Indian context have been done so in Hindi language. Very few instruments are available in other languages. The scales/instruments that are available in multiple languages include Mini International Neuropsychiatric Interview (MINI), General Health Questionnaire (GHQ), Patient Health Questionnaire (PHQ), Montreal Cognitive Assessment (MoCA), Geriatric Depression Scale (GDS), Edinburgh postnatal depression, Epworth sleepiness scale to evaluate daytime sleepiness, Columbia-Suicide Severity Rating Scale (C-SSRS), Recovery quality of life, World Health Organization Quality of Life-Bref version, Subjective happiness scale, Hospital Anxiety and Depression scale (HADS), Perceived Stress Scale (PSS), Multidimensional Scale of Perceived Social Support (MSPSS), Internalized Stigma of Mental Illness (ISMI), COVID-19 stigma scale, Dyadic adjustment scale, Broad Autism Phenotype Questionnaire (BAPQ), Strength and Difficulties Questionnaire (SDQ), and Rosenzweig picture frustration study (Children's form). Overall, very few instruments have been developed in India. To conclude, our review suggests that a limited number of scales have been developed in India and in terms of translation and adaptation, most of these have been done in Hindi only. Keeping these points in mind, there is a need to develop more psychometrically sound scales for research and routine clinical practice. Additionally, efforts must be made to translate and adapt scales available in different languages and subject the same to psychometric evaluations.

Linguistic borrowings in trade terminologies: an analysis of ancient Indian and Egyptian languages from 3300 BCE to 500 CE

The research delves into the intricate web of linguistic borrowings in trade terminologies between Ancient Indian languages (Sanskrit, Prakrit) and Egyptian languages (Ancient Egyptian Hieroglyphs, Demotic script) spanning from approximately 3300 BCE to 500 CE. Through a meticulous analysis of selected literary texts and inscriptions, including the Hathigumpha Inscription, Rudradaman I Inscription, Junagadh Rock Inscriptions, Nasik Cave Inscriptions, Rosetta Stone, Satirical Papyrus, and Turin Taxation Papyrus, the study unravels the linguistic evolution and cultural exchanges embedded within ancient trade networks. Contextualizing these linguistic borrowings within the historical framework of trade routes, maritime activities, and economic exchanges, the research sheds light on the nuances of communication and adaptation in commercial interactions. By employing rigorous research methodologies and comparative studies, the findings offer a comprehensive understanding of how language served as a bridge for commerce, diplomacy, and cultural exchange between these ancient civilizations. This study not only contributes to the fields of linguistic history and ancient trade studies but also offers valuable insights into the dynamic interplay of language, trade, and cultural connectivity in the ancient world.

Interlingual Syntactic Parsing: An Optimized Head-Driven Parsing for English to Indian Language Machine Translation

In the era of Artificial Intelligence (AI), significant progress has been made by enabling machines to understand and communicate in human languages. Central to this progress are parsers, which play a vital role in syntactic analysis and support various Natural language Processing (NLP) applications, including Machine Translation and sentiment analysis. This paper introduces a robust implementation of an optimized Head-Driven Parser designed to advance NLP capabilities beyond the limitations of traditional Lexicalized Tree Adjoining Grammar (L-TAG) based Parser. Traditional parser, while effective, often struggle with the capturing complexities of natural languages, especially translation between English to Indian languages. By leveraging Bi-directional approach and Head-Driven techniques, this research offers a revolutionary enhancement in parsing frameworks. This method not only improves performance in syntactic analysis but also facilitates complex tasks such as discourse analysis and semantic parsing. This research involves experimentation the Bi-Directional Parser on a dataset of 15,000 sentences, resulting a reduction in derivation variations compared to conventional TAG Parsers. This advancement highlights how Head-Driven Parsing can overcome traditional constraints and provide more reliable linguistic analysis. The paper demonstrates how this new implementation not only builds on the strengths of L-TAG but also addresses its limitations and contributes to expanding the scope of Tree Adjoining Grammarbased methodologies and advancing the field of Machine Translation.

Hate speech detection in low-resourced Indian languages: An analysis of transformer-based monolingual and multilingual models with cross-lingual experiments

Abstract Warning: This paper is based on hate speech detection and may contain examples of abusive/ offensive phrases. Cyberbullying, online harassment, etc., via offensive comments are pervasive across different social media platforms like ™Twitter, ™Facebook, ™YouTube, etc. Hateful comments must be detected and eradicated to prevent harassment and violence on social media. In the Natural Language Processing (NLP) domain, the most prevalent task is comment classification, which is challenging, and language models based on transformers are at the forefront of this advancement. This paper intends to analyze the performance of language models based on transformers like BERT, ALBERT, RoBERTa, and DistilBERT on the Indian hate speech datasets over binary classification. Here, we utilize the existing datasets, i.e., HASOC (Hindi and Marathi) and HS-Bangla. So, we evaluate several multilingual language models like MuRIL-BERT, XLM-RoBERTa, etc., few monolingual language models like RoBERTa-Hindi, Maha-BERT (Marathi), Bangla-BERT (Bangla), Assamese-BERT (Assamese), etc., and perform cross-lingual experiment also. For further analyses, we perform multilingual, monolingual, and cross-lingual experiments on our Hate Speech Assamese (HS-Assamese) (Indo-Aryan language family) and Hate Speech Bodo (HS-Bodo) (Sino-Tibetan language family) dataset (HS dataset version 2) also and achieved a promising result. The motivation of the cross-lingual experiment is to encourage researchers to learn about the power of the transformer. Note that no pre-trained language models are currently available for Bodo or any other Sino-Tibetan languages.

Pilot Testing and Vernacular Translation of EORTC Patient Satisfaction Questionnaire (PATSAT-C33 and OUT-PATSAT7) at a Tertiary Care Cancer Center in India

Purpose Pilot testing and translation of the English version of European Organization for Research and Treatment of Cancer (EORTC) patient satisfaction cancer core questionnaire (PATSAT-C33) and complementary outpatient module (OUT-PATSAT7) into two Indian vernacular languages (Hindi and Marathi). Methods Patients undergoing fractionated radiotherapy for cancer with basic proficiency in respective language were included in the study after written informed consent. The English version of EORTC PATSAT-C33 and OUT-PATSAT7 questionnaire was pilot tested in 20 patients. The questionnaire was then translated into two Indian vernacular languages (Hindi and Marathi) using EORTC translation methodology. This included forward-translation by two independent professional translators into target languages (Hindi and Marathi) to create an intermediate version; back-translation into English by another independent pair of linguistic experts; and harmonization by comparing back-translated versions (English) to the original English version for reconciliation. The EORTC translation group provided suggestions and proofread the reconciliated versions (Hindi and Marathi) which were then administered to 20 patients in each language. Semistructured interviews were conducted for patients to identify problems in understanding the translation versions to make appropriate corrections/modifications to the questionnaire. Results Pilot testing of English version of PATSAT-C33 and OUT-PATSAT7 did not pose any major difficulty leading to subsequent translation into both target languages (Hindi and Marathi). Reconciliated version of the translated questionnaires was arrived at after incorporating suggestions and proofreading by the EORTC translation group. Pilot testing of the reconciliated questionnaires (Hindi and Marathi) did not identify major problems in understanding, difficult/confusing words, or upsetting questions leading to the adoption of the reconciliated version as final translated questionnaire without further modifications. Conclusion The English version of PATSAT-C33 and OUT-PATSAT7 has been successfully translated into Hindi and Marathi languages using standardized EORTC methodology. Psychometric properties of the same are currently being tested for validation in a larger Indian cohort.