Kannada Research Papers - Academia.edu (original) (raw)

Top Papers
Most Cited Papers
Most Downloaded Papers
Newest Papers
People
- by
- •
- Hinduism, Kannada, Medieval Indian History, Mediaeval Deccan

We report our recent work on the recognition of scene text captured by mobile cameras, which we have named Kannada Pado. The text region is currently manually cropped using a user-friendly interface, which permits repeated croppings from the captured image in a hierarchical fashion. The scene text segment is then binarized using the algorithm, midline analysis, and propagation for segmentation. The segmented binary text image is recognized using Lipi Gnani Kannada OCR. The recognized text can be transcribed in Roman, Devanagari, and other principal Indian scripts. Such tools will be of immense use in metropolitan cities such as Bengaluru for business visitors and tourists to be able to read important textual information using their mobile itself. The entire implementation is of low computational complexity and hence, runs fully on the mobile itself, without any backend computation. Currently, text recognition accuracy is the bottleneck, which, when improved, will make the app immediately usable by people. Then, it will be made available to the public from Google Playstore.

- by Sasidharan Ayyavu
- •
- Machine Learning, Pattern Recognition, Kannada, Tamil

The status of basic grammatical relations such as 'subject-of'has been a matter of some controversy in current linguistic theory. While Keenan has argued that they are squishy of 'cluster'concepts, Johnson and others have taken the position that they must be analyzed as discrete entities. This paper provides empirical evidence to decide between the two opposing positions. The syntax and semantics of the so-called Dative Subject Construction in South Asian languages demonstrate that the underlying subject is different ...

- by S.N. Sridhar
- •
- Cognitive Science, Languages and Linguistics, Indian studies, South Asian Studies

This is an Introduction to the series of texts translated from Kannada into English which I was Coordinating.

- by arpitha rao
- •
- Languages and Linguistics, Kannada, Consonant Clusters, Minimal Pair

Titled "Niraakarana" in Kannada, this poem by Roopa Hassan - one of the major contemporary Kannada women poets - beautifully voices the concerns and precepts of the subaltern whose source of compassion is born out of their suffering. Here... more

Social media plays an inevitable role in the life of millions of people worldwide. Social media platforms are increasingly being used for various purposes in a wide range of fields, which also includes biodiversity conservation. Evidently, biodiversity has been providing the base for livelihoods, cultures and economies of several hundred millions of people, including farmers, fisher folk, forest dwellers and artisans. This immense resource of biodiversity must be conserved for the good of today and tomorrow. Conservation activities can be initiated through awareness campaigns among the public about the role and importance of biodiversity in keeping the balanced ecosystem. Technological development in the field of Media has lead to the invention of Internet and Social media platforms. As of 2017 Statista report (International statistics database), India stands second among Internet users with 462 millions and by 2019 the social networks users number estimated to reach 258 millions. Its more than half of today’s Internet users number. These huge numbers itself reveals the potentiality of social media as a strategic tool in enhancing the awareness biodiversity conservation. This paper attempts to explore the awareness levels of biodiversity and its conservation ideas among youth with the help of empirical as well as secondary data.

- by Vinay G P and +1
- •
- Media Studies, Citizen Journalism, Social Media, Kannada

Kannada is major Dravidian language spoken by about seven crore people in the state of Karnataka in South India. T Burrow and M B Emeneau were the first to bring out an etymological dictionary for the Dravidian languages amongst which Kannada is one. Then on a tradition of such works is built in the language. Here is a depiction of this tradition.

- by Maheedas Bv
- •
- Kannada, Book Reviews, Language and Etymology
- by Caleb Simmons
- •
- South Asian Studies, South Asian History, Kannada, Kannada language and literature

In this chapter, I attempt to provide a different a way of viewing goddesses by examining how they develop through their roles in the legitimation of kingship in South India and how that relates to the Purāṇic conception of the cosmos. My aim is to demonstrate one process through which a local goddess has, over a period of time, become associated with pan-Indian Purāṇic deities yet retained a distinctly local flavor. In the medieval period during which all kings of South India were configured within complex of hierarchic relationships based on a Purāṇic model of overlordship, the imagined separation between “vernacular” and “Sanskritic” or “Brahminic” only masks the significant role of goddesses at various stages in the construction of statehood. Their function in the construction of polity provides more insight into their place in the study of medieval Indian society than those that arise through a discussion of the “high Sanskritic” tradition or the “low vernaculars,” especially since both regional and local kingdoms operated according to the model of statehood provided in Purāṇic material.

- by Caleb Simmons
- •
- Religion, History, History of Religion, Kannada

Translation of a Short Story from Kannada written by Fakir Mohmad Katpadi, Translated by Tharakeshwar V.B.

ಸುಮಾರು ಎರಡು ಸಾವಿರ ವರ್ಷಗಳ ಇತಿಹಾಸ ಹೊಂದಿರುವ ನಮ್ಮ ಕನ್ನಡ ಭಾಷೆಯು ಅತೀ ಶ್ರೀಮಂತವೂ, ಬಹು ವಿಚಾರ ಆಗರವೂ, ದಾಸ ಶರಣರ ಸಾಹಿತ್ಯಗಳಿಂದ ಅಜರಾಮರವೂ ಆಗಿರುತ್ತದೆ. ಈ ಶ್ರೇಷ್ಠ, ಸರಳ ಸುಂದರ ಸುಮಧುರವಾದ ಕನ್ನಡ ಭಾಷೆಯು ಅನೇಕ ಪರಕೀಯರ ದಾಳಿಗಳನ್ನೆಲ್ಲಾ ಸಹಿಸಿಕೊಂಡು ತನ್ನ ಗರಿಮೆಯನ್ನು ಸ್ವಾಸ್ಥ್ಯವಾಗಿ ಹಿಡಿದಿಟ್ಟುಕೊಂಡಿದ್ದು ನಮ್ಮ ಕನ್ನಡಿಗರ ಹಿರಿಮೆಯಾಗಿದೆ. ಕನ್ನಡಾಂಬೆಯ ರಕ್ಷಣೆಗೆಂದೇ ಸದಾಕಾಲ ತಮ್ಮ ಜೀವನವನ್ನೇ ವ್ಯಯಿಸುತ್ತಿರುವ ನಮ್ಮ ಕವಿ – ಕುಲ - ಸಾಹಿತಿಗಳ ಪರಂಪರೆಯು ಆ ತಾಯಿಯ ಹೆಮ್ಮೆಯ ಸೇವಕರೂ, ರಕ್ಷಕರೂ ಆಗಿದ್ದಾರೆ. ಆದರೂ ಇಂದು ನಮ್ಮ ಕನ್ನಡ ಕಲಿಕೆಯಲ್ಲಿ ಆಗುತ್ತಿರುವ ಧೋರಣೆ ಬಹಳ ಬೇಸರವನ್ನುಂಟು ಮಾಡುತ್ತಿದೆ. ಗಿಡವಾಗಿ ಬಗ್ಗದ್ದು ಮರವಾಗಿ ಬಗ್ಗೀತೆ? ಎಂಬ ಗಾದೆಯ ಹಾಗೆ ನಮ್ಮ ಮಕ್ಕಳು ಪ್ರಾರಂಭದಿಂದ ಅಭ್ಯಾಸಿಸಬೇಕಾದ ಕನ್ನಡ ತಾಯ್ನುಡಿಯನ್ನು, ಇನ್ನಾವಾಗ ಕಲಿತಾರು? ಇದಕ್ಕೆ ಪುಷ್ಠಿ ನೀಡುವಂತೆ ಆಂಗ್ಲ ಮಾಧ್ಯಮವು, ಸ್ವತಃ ತಿದ್ದಬೇಕಾದ ತಂದೆ ತಾಯಿರನ್ನೂ, ಪೋಷಕರನ್ನೂ ಸೆಳೆಯುತ್ತಿರುವದು ಅತೀ ನೋವಿನ ಸಂಗತಿಯಾಗಿದೆ.

- by Subrahmanya G M
- •
- Education, Kannada, Kannada language and literature

The language variety of the Hebbar Iyengar community of Karnataka (henceforth ‘Hebbar’) is a minority Dravidian language variety (no reliable statistics; unofficial estimates ~ 100,000 speakers) spoken in the areas of southern Karnataka... more

- by Arvind Iyengar
- •
- Phonology, Dialectology, Kannada, Tamil

We investigated children's spelling in the alphasyllabic writing system of Kannada, a South Indian language. Kannada represents language at roughly the level of the syllable, but its symbols or 'akshara' can be segmented visually to reveal the consonant and vowel segments within. Having been taught akshara as whole, fourth-and fifth-grade children were poor at constructing complex akshara from their consonant and vowel segments. Analyses of their errors showed confusions between the symbols for phonologically similar consonants, visually based errors, particularly on small diacritic marks, and influences of dialect. Kannada has been considered a transparent writing system, but the challenges that arise in mastering this visuo-spatially complex and extensive orthography suggest that previous studies have missed certain factors that contribute to the ease or difficulty with which a writing system is learned.

- by Rebecca Treiman and +1
- •
- Languages, Psychology, Cognitive Psychology, Languages and Linguistics

Singing the Goddess into Place examines Chamundi of the Hill, a collection of songs that tells the stories of the gods and goddesses of the region around the city of Mysore in southern Karnataka. The ballad actively transforms the region into a land where gods and goddesses live, embedding these deities within the social worlds of their devotees and remapping southern Karnataka into sacred geography connected through networks of devotion and pilgrimage. In this in-depth study of the songs and their context, Caleb Simmons not only provides the first English-language translation of these songs but brings to light the unstudied folk perspectives on the foundational myth of Mysore and its urban history. Singing the Goddess into Place demonstrates how folk narratives reflect local context while also actively working to upend social inequities based on caste and ritual/devotional practices. By delving into this world, the book helps us understand how a landscape is transformed through people's relationship with it and how this relationship helps build meaning for the communities that call it home.

- by Caleb Simmons
- •
- Hinduism, Folklore, Southeast Asian Studies, Global Social Change

We propose data and knowledge-driven approaches for multilingual training of the automated speech recognition (ASR) system for a target language by pooling speech data from multiple source languages. Exploiting the acoustic similarities between Indian languages, we implement two approaches. In phone/senone mapping, deep neural network (DNN) learns non-linear functions to map senones or phones from one language to the others, and the transcriptions of the source languages are modified such that they can be used along with the target language data to train and fine-tune the target language ASR system. In the other approach, we model the acoustic information for all the languages simultaneously by training a multitask DNN (MTDNN) to predict the senones of each language in different output layers. The cross-entropy loss function and the weight update procedure are modified such that only the shared layers and the output layer responsible for predicting the senone classes of a language are updated during training, if the feature vector belongs to that particular language. In the low-resourced setting (LRS), 40 hours of transcribed speech data each for Tamil, Telugu and Gujarati languages are used for training. The DNN based senone mapping technique gives relative improvements in word error rates (WERs) of 9.66%, 7.2% and 15.21% over the baseline system for Tamil, Gujarati and Telugu languages, respectively. In medium-resourced setting (MRS), 160, 275 and 135 hours of data for Tamil, Kannada and Hindi languages are used, where, the same technique gives better relative improvements of 13.94%, 10.28% and 27.24% for Tamil, Kannada and Hindi, respectively. The MTDNN with senone mapping based training in LRS, gives higher relative WER improvements of 15.0%, 17.54% and 16.06%, respectively for Tamil, Gujarati and Telugu, whereas in MRS, we see improvements of 21.24% 21.05% and 30.17% for Tamil, Kannada and Hindi languages, respectively. CCS Concepts: • Computing methodologies → Speech recognition.

- by Ramakrishnan Angarai Ganesan
- •
- Automatic Speech Recognition, Speech Recognition, Telugu, Kannada

To theorize Jain sovereignty, this essay takes up Ernst Kantorowicz’s underlying query of what happens when a king dies. In turning to medieval Jain authors such as Jinasena, we see how sovereignty and renunciation were mutually constituted such that the king’s renunciation completely subverts the problem of the king’s death. If the fiction of Jain kingship properly practiced culminates in renunciation, then such a movement yields up a new figure of the ascetic self-sovereign. Renunciation does not sever sovereignty but extends it into a higher spiritual domain. Worldly and spiritual sovereignty share a metaphorical language and set of techniques that render them as adjacent but hierarchical spheres of authority. In so doing, Jain authors provide a religious answer to a political problem and make the political inbuilt into the religious, thereby revealing their interpenetrating and bounded nature.

- by Sarah Pierce Taylor
- •
- Jainism, Kannada, Digambara Authors, Kannada language and literature

Etymological dictionary of Kannada language.

- by ravi vararo
- •
- Etymology, Kannada, kannada dictionary

We participated in the Blizzard Challenge 2013 for Tamil and Kannada, using our unit selection based concatenative speech synthesis system. Sentence level viterbi search is used to select the reliable speech units among a set of candidate units. The same Text-to-Speech synthesis framework is used to synthesize speech in both the Indian languages. The given Wikipedia and semantically unpredictable test sentences are synthesized using IIIT-H Indic corpus and the listening test results reported by the blizzard evaluation team is discussed. The letter code for MILE TTS is “R”.

- by Rajaram Belakawadi and +1
- •
- Human Computer Interaction, Speech Synthesis, Kannada, Tamil
- by S. N. Sridhar
- •
- Languages and Linguistics, Indian studies, South Asian Studies, Syntax

In this paper, we propose a novel dexterous technique for fast and accurate recognition of online handwritten Kannada and Tamil characters. Based on the primary classifier output and prior knowledge, the best classifier is chosen from a set of three classifiers for second-stage classification. Prior knowledge is obtained through analysis of the confusion matrix of primary classifier which helped in identifying multiple sets of confused characters. Further, studies were carried out to check the performance of secondary classifiers in disambiguating among the confusion sets. Using this technique we have achieved an average accuracy of 92.6 % for Kannada characters on the MILE lab dataset and 90.2 % for Tamil characters on the HP Labs dataset.

- by Sarah Pierce Taylor
- •
- Kannada, Kannada language and literature, Karnataka, Bhakti

Shrti Shringara ia an anthology of poems collected from my earlier collections of poems. There are fifty poems in this collection. Some of these are translations from classical authors and I have indicated the names of the original authors at the end of such poems.
These I think will interest the readers.
I thank heartily Mr Varuna for having helped me in bringing out this book in this form and also designing the cover page.

- by Maheedas Bv
- •
- Kannada, Kannada language and literature, Kannada poems

This paper presents the design of a full fledged OCR system for printed Kannada text. The machine recognition of Kannada characters is dificult due to similarity in the shapes of different characters, script complexity and non-uniqueness in the representation of diacritics. The document image is subject to line segmentation, word segmentation and zone detection. From the zonal information, base characters, vowel modifiers and consonant conjuncts are separated. Knowledge based approach is employed for recognizing the base characters. Various features are employed for recognising the characters. These include the coefficients of the Discrete Cosine Transform, Discrete Wavelet Transform and Karhunen-Louve Transform. These features are fed to different classifiers. Structural features are used in the subsequent levels to discriminate confused characters. Use of structural features, increases recognition rate from 93% to 98%. Apart from the classical pattern classification technique of nearest neighbour, Artificial Neural Network (ANN) based classifiers like Back Propagation and Radial Basis Function (RBF) Networks have also been studied. The ANN classifiers are trained in supervised mode using the transform features. Highest recognition rate of 99% is obtained with RBF using second level approximation coefficients of Haar wavelets as the features on presegmented base characters.

- by Ramakrishnan Angarai Ganesan and +1
- •
- Artificial Intelligence, Image Processing, Natural Language Processing, Machine Learning

In this article, we address the challenges in segmentation of online handwritten, isolated Kannada words. This is the maiden work in the field of segmentation of online handwritten Kannada words. Due to the advent of tablet PCs and systems with pen-enabled interface, online handwriting recognition has got wide applications such as form filling, field data collection and word processing. In some of these applications, recognition of names of individuals is required and it is nearly impossible to maintain the lexicon of all possible names. Also, Kannada, being a Dravidian language, is morphologically rich and also agglutinative and thus does not have a finite lexicon. For example, a single root verb can easily lead to a few thousand words after morphological changes and agglutination. Hence, to make recognition of open vocabulary online handwritten Kannada words possible, one must necessarily look into the possibility of segmentation of online Kannada words into their constituent symbols.
The modern Kannada script consists of 13 vowels, 34 consonants, 2 other symbols, and 10 numerals. Old Kannada script has additional 3 vowels and 2 consonants. In Kannada, a compound character called akshara is possible by combining consonants and vowels. A total of 8,63,848 different aksharas are possible. We derived a minimal set of 295 distinct symbols to recognize these characters. A corpus of isolated Kannada symbols (MILE Lab Kannada symbols data-set) is collected to study the various statistics about Kannada characters. For addressing the issue of segmentation, Kannada words are collected using a custom
application running on a tablet PC from high school. We refer to this dataset as MILE Lab Kannada word dataset.
We address the challenges in any online Kannada word into its constituent symbols. We approach the segmentation problem in two steps: preliminary segmentation (PS) and attention feed-based segmentation (AFS). In PS, we use the heuristics from language constructs to segment the word into a set of stroke groups (SG). Then in AFS method, we extract some features from each SG to suspect any under-segmentation or over-segmentation. Then we make use of (i) probability estimate from the classifier and (ii) attributes such as number of dominant points and inter stroke displacements from prior knowledge, to resolve the suspected stroke group by splitting or regrouping or retaining it. A segmentation accuracy of 83.4% is achieved by PS, which is increased to 94.3% by AFS.
We use SVM classifier to recognize and get probability estimates for individual stroke groups. SVM is trained on MILE Lab Kannada symbols training data-set. The smoothed, normalized and re-sampled x, y-coordinates of online trace are used as a feature vector for each sample in the data-set. The character recognition accuracy of 56% is achieved by recognizing the PS outputs and 62% by recognizing the symbols segmented by AFS on the MILE Lab Kannada words.

- by Ramakrishnan Angarai Ganesan and +1
- •
- Human Computer Interaction, Machine Learning, Classification (Machine Learning), Pattern Recognition
- by Sonali Nag
- •
- Psychology, Cognitive Psychology, Languages and Linguistics, Phonology

Reading and writing difficulties are markers for some forms of learning disorders, and measuring the distance between the child’s performance and an expected level of attainment is a common approach to diag-nosis. However, there are several problems with relying on the gap between achievement and expectation for arriving at a diagnosis, not least because the approach is essentially blind to the child’s history of opportunity. An alternative approach offers children tiers of individual-ised programmes, and their changing strengths and weaknesses informs the diagnosis. This article reports the findings of an exploratory study analysing the change scores of ten children who were reading two years behind their class level. We offered a two-tiered language intervention, embedded within a whole class programme, using the children’s narra-tive writing for examining change. Our findings suggest that the profile of strengths and weaknesses gathered from a school-based intervention is a valua...

- by Sonali Nag
- •
- Psychology, Kannada, Dyslexia

Our nation is exceptionally rich in linguistic diversity and relatively poor in modern scientific temper. Therefore, dissemination and innovation of scientific knowledge in Indian languages is an imperative for the modernisation and development of Indian society. However, dissemination and innovation of scientific knowledge in Indian languages is limited by two impediments. First is the absence of adequate linguistic skills of science communication in Indian languages. Second is the limited language resources for science communication in Indian languages. Unlike the earlier generation, the skill of the communication of science in Indian languages is not up to the increasing requirements among the new generation.

- by P Sreekumar
- •
- Sinhala and Dravidian languages, Malayalam, Telugu, Kannada
- by P Arulmozhi
- •
- Human Computer Interaction, Signal Processing, Speech Synthesis, Prediction
- by S.N. Sridhar
- •
- Cognitive Science, Languages and Linguistics, Indian studies, South Asian Studies

Advertorial is a combination of advertisements and editorial content. Editorial space is sold and used for advertorials, promotional news/articles, any form of paid news. All forms of media are susceptible to this kind of promotion, especially regional newspapers and TV Channels with large readership/viewership and play a major role in moulding public opinion often. Paid Promotional News is the new practice applied by the advertisers, their clients, Govt., Institutions, and Political Parties who gain substantially through favourable public opinion by portraying themselves in positive light. With this emerging trend it is necessary to study the effect of advertorials on the readers and to know whether readers realize that these are paid news or not. As no such study was conducted on Kannada newspapers, this study tries to put-forth an clear picture of credibility of advertorials and their effect on the credibility of the newspapers carrying them. In this study, multistage sampling method, appropriate statistical analysis for interpretation and agenda setting theory were applied. This study shows that the promotional news/articles are not as credible as regular news/articles.

- by Abhilasha R
- •
- Mass-Media Ethics, Kannada, Advertorials

In interpreting a sentence, listeners rely on a variety of linguistic cues to assign grammatical roles such as agent and patient. The present study considered the relative ranking of three cues to agenthood (word order, noun animacy, and subject-verb agreement) in normal and aphasic speakers of Hindi. Because animacy plays a grammatical role in Hindi (determining the nature and acceptability of sentences without accusative marking), this language is relevant to the claim that Broca's aphasia involves a dissociation between grammar and semantics. Results of Study 1 with normal Hindi-dominant speakers showed that animacy is the strongest cue in this language, while agreement is the weakest cue. In Study 2, Hindi-English bilinguals were tested in both their languages. Most showed the normal animacy-dominant monolingual pattern in Hindi, with a mixture of strategies from both languages in their interpretation of English. A substantial minority showed mixed strategies in both languages. Only 5 of 48 subjects displayed a complete separation between languages, with animacy dominance in Hindi and word order dominance in English. In Study 3, two Hindi-English bilinguals with Broca's aphasia were tested in both languages. Results indicate (a) greater use of animacy in Hindi than in English and (b) greater use of word order in English than in Hindi. The strategies displayed by these patients fall well within the range observed among bilingual normals. We conclude that the use of animacy in sentence interpretation by these aphasic patients reflects preservation of normal, language-specific processing strategies; it cannot be interpreted as a nonlinguistic strategy developed to compensate for receptive agrammatism. Results are discussed in light of other cross-linguistic evidence on sentence comprehension in monolingual and bilingual aphasics.

AbstractIn this paper, we present an unrestricted Kannada online handwritten character recognizer which is viable for real time applications. It handles Kannada and Indo-Arabic numerals, punctuation marks and special symbols like $, &, #... more

The study was conducted in Honnaver taluka (14° 8′ 0″ N to 14° 25′ N Latitude and 74° 25′ E to 74° 45′ E Longitude) of Uttar Kannada district, Karnataka India to assess the carbon sequestration in soils of different land use systems. The IRS P6 LISS-III imageries of the study area was procured from NRSC, Hyderabad and different land use systems in Honnaver taluka were identified with the ground truth data collected from GPS and processed in ERDAS software. The land use land cover (LULC) classes, viz., dense forest, sparse forest, agriculture and open land were identified. The total area in each class was assessed through supervise classification. The soil samples at 1 m depth were drawn at grid point in flat land and along the profile in sloppy land in different land use system. The SOC was estimated using Walkley and Black rapid titration method. The total area in four land use classes is 68520 ha with SOC pool of 12.112 million tonnes. Among different classes dense forest covers h...

- by Pavan Tikhile
- •
- Environmental Science, Kannada, Environmental Science and Engineering

- by Gangotri Nadiger
- •
- Machine Learning, Pattern Recognition, Kannada, Tamil

Klatt synthesizers are one of the most widely used formant synthesizers. Klatt synthesizers are usually implemented either with fixed parameter update rate or with variable parameter update rate. This paper proposes new method to store control parameters and parameter update strategy to improve the naturalness of the synthesized vowel and semi vowel sounds.

Describes A knowledge based approach and implementation to designing a word processor for Kannada and Indian languages in general.

Medical Intelligence and Language Engineering Laboratory at the Indian Institute of Science developed Kannada and Tamil OCR and TTS systems and deploying them, created Braille and audio books for blind students and won Manthan Awards. Algorithms proposed for scene and born-digital word image recognition consistently retained the top positions in ICDAR Robust Reading Competitions since 2011. The TTS systems were adjudged second best in Blizzard TTS Challenge in 2013 and 2014. Current research is on solving the problems of unlimited vocabulary in building ASR systems for Dravidian languages of Kannada and Tamil, by using sub-word units for language modeling.