We have identified Tamil natural language processing research projects that should be initiated either by building upon previous work, towards creating systems with better performance and evaluation measures or building systems using new machine learning approaches. Meanwhile we also want to do a complete linguistic study on the evolution of the Tamil language. The projects are organised in to six main research themes and covers all areas related to natural language processing for Tamil.
Language Resources
Monolingual/Parallel/Annotated Text/Speech Corpora
Language resources form the foundation of all natural language processing research and applications. From dictionary development and language modelling to machine translation, speech technologies, and generative AI systems, the availability of large-scale, high-quality corpora is essential for technological advancement. A corpus is a structured collection of textual and spoken language data drawn from diverse sources such as newspapers, books, magazines, academic publications, literary works, web content, social media, and archival collections.
We aim to develop comprehensive monolingual corpora for Tamil, Sinhala, and other Dravidian languages, as well as parallel corpora for Tamil-English, Tamil-Sinhala, and Tamil-Dravidian language pairs. These resources will include annotated corpora enriched with linguistic information such as Part-of-Speech tags, morphology, lemmas, syntactic structures, named entities, semantic relationships, and discourse information. In addition, we aim to create speech corpora that support automatic speech recognition, speaker recognition, speech synthesis, and conversational AI applications.
Special emphasis will be placed on the creation of dedicated resources for Sri Lankan Tamil. Although Tamil is a shared language across multiple regions, significant linguistic variations exist in vocabulary, pronunciation, syntax, and usage. Building separate lexical, textual, and speech resources for Sri Lankan Tamil is therefore an important objective. These resources will support the development of language technologies specifically designed for Sri Lankan Tamil-speaking communities.
Furthermore, we seek to expand traditional corpus development by incorporating large-scale digital archives, historical newspapers, manuscripts, web archives, and born-digital content. Such resources will enable not only language technology development but also research into cultural heritage, historical discourse, and language evolution.
Language/Speech Models
Language and speech models are fundamental components of modern natural language processing and speech technologies. They enable applications such as machine translation, speech recognition, speech synthesis, information retrieval, question answering, conversational AI, and text generation. Traditional statistical approaches remain valuable for modelling language patterns, while recent advances in machine learning have enabled the development of large-scale neural and foundation models capable of capturing complex linguistic and semantic relationships.
We aim to develop efficient language models for Tamil using statistical, neural, and morpheme-based approaches. Statistical language models assign probabilities to sequences of words based on observed language usage, while morpheme-based models are particularly useful for highly inflectional languages such as Tamil, where words may consist of multiple grammatical units. By modelling stems, affixes, and morphological structures explicitly, such systems can better capture linguistic variation and improve performance in downstream applications.
In addition to traditional language models, we seek to develop foundation models trained on large-scale Tamil corpora representing both contemporary and historical language usage. These models will support applications such as semantic search, summarization, machine translation, conversational AI, and knowledge discovery.
For speech technologies, we aim to develop robust acoustic and speech models trained on diverse spoken Tamil datasets. These models will support automatic speech recognition, speaker identification, speech synthesis, and voice-based human-computer interaction. Particular attention will be given to dialectal diversity, including Sri Lankan Tamil and regional spoken varieties, ensuring broad applicability across communities.
Dictionaries
Lexical resources provide essential linguistic knowledge for both human users and computational systems. Traditional dictionaries contain information about pronunciation, meaning, etymology, grammatical categories, and usage patterns, while computational dictionaries provide structured lexical information for language processing applications.
We aim to collect, digitize, standardize, and expand existing lexical resources for Tamil. These resources include monolingual and bilingual dictionaries, thesauri, pronunciation dictionaries, terminology databases, and domain-specific lexicons. Where possible, we will collaborate with academic institutions, linguistic experts, and cultural organizations to preserve and enhance existing resources.
Beyond conventional dictionaries, we aim to develop computational lexical resources that represent semantic relationships, conceptual hierarchies, and linguistic knowledge in machine-readable formats. These resources will serve as foundational infrastructure for machine translation, information retrieval, language modelling, speech technologies, and AI-based knowledge systems.
WordNet for Tamil
WordNet is a lexical-semantic network that organizes words into sets of synonyms known as synsets, representing shared concepts and meanings. These synsets are interconnected through semantic relationships such as synonymy, antonymy, hypernymy, meronymy, and conceptual association. WordNet has become a foundational resource for computational linguistics and supports applications including machine translation, information extraction, word sense disambiguation, knowledge representation, semantic search, and question answering.
We aim to expand and modernize WordNet resources for Tamil by capturing lexical relationships, concepts, meanings, and semantic structures across diverse domains. This includes identifying root words, lexical senses, semantic categories, and conceptual relationships, and organizing them into an accessible and scalable knowledge base.
In addition to traditional WordNet development, we seek to integrate lexical-semantic resources with ontologies, taxonomies, and knowledge graphs. This will enable the creation of richer semantic knowledge networks that support advanced AI applications, knowledge discovery, cultural heritage preservation, and intelligent information access systems. The resulting resources will be made available through online platforms and open research interfaces, encouraging collaboration among researchers, institutions, and the broader Tamil language community.
Language Parsing and Resolution
Tokenizer
Tokenization is the process of breaking a stream of text up into words, phrases, symbols, or other meaningful elements called tokens. The list of tokens becomes input for further processing such as parsing or text mining. Tamil uses spaces to mark word boundaries, but a lot of Tamil word forms are agglutinative in nature, meaning they glue together at least two words. Those cases can be identified as determiners+nouns, nouns+postpositions, verbs+particles, nouns+particles and etc. Except the first pattern (determiners+nouns), in all other cases, the second part of the word forms are restricted and can be listed. So it is possible to split certain Tamil agglutinative word forms into separate tokens. This project would aim to build a module that splits Tamil sentences into words carefully studying the different word forms appearing in sentences.
Chunker
Chunking is the task of identifying and segmenting the text into syntactically correlated word groups. It could divide a sentence into its major non-overlapping phrases and attach a label to each chunk. Tamil being an agglutinative language have a complex morphological and syntactical structure. It is a relatively free word order language but in the phrasal and clausal construction it behaves like a fixed word order language. In Tamil language Transformation Based Learning and Machine learning techniques using SVMs have been used in building a Chunker. Chunkers could be based on Noun phrases, Verb Phrases, Adjectival Phrases, Adverbial Phrases, Conjunctions and Complementizers. This project would aim to build a module that performs chunking in the above mentioned categories when given Tamil text.
Sentence Splitter
Sentence boundary disambiguation (SBD), also known as sentence breaking, is the problem in natural language processing of deciding where sentences begin and end. Often natural language processing tools require their input to be divided into sentences for a number of reasons. However sentence boundary identification is challenging because punctuation marks are often ambiguous. For example, a period may denote an abbreviation, decimal point, an ellipsis, or an email address and not the end of a sentence. also, question marks and exclamation marks may appear in embedded quotations, emoticons and slang. The task of sentence splitting is quite important for natural language processing tasks such as machine translation, speech recognition and language modelling. This project would aim to build an efficient sentence splitter module that could be used to break down Tamil text in to meaningful sentences.
Part-of-Speech Tagger
Part of speech (POS) tagging is the process of labeling a part of speech or other lexical class marker to each and every word in a sentence. Tamil being a Dravidian language has a very rich morphological structure which is agglutinative. Tamil words are made up of lexical roots followed by one or more affixes and therefore tagging a word on Tamil is quite complex. A morphologically rich language such as Tamil has many grammatical categories which leads to ambiguity. Hence lexical ambiguity must be taken into consideration during the development of an efficient POS tagger. In the past Rule based methods, hidden mark models and support vector machines have been used for developing a POS tagger for Tamil. This project will aim to build a Tamil POS tagger module that addresses existing challenges and performs with high accuracy.
Grammatical Parser / Dependency Parser
Dependency grammar is based on the idea that the syntactic structure of a sentence consists of binary asymmetrical relations between the words of the sentence. Dependency parsing uses linguistic information to give relationship between words. In the past only few attempts have been reported regarding dependency parsing for Tamil. They have used rule-based and corpus-based approaches in building a dependency parser. An annotation scheme needs to be designed in order to annotate Tamil data with dependency relations to train a dependency parser and this is a very challenging task. A series of linguistic rules have to be implemented to build dependency structure for Tamil sentences. This project aims to build a dependency parsing module for Tamil which could identify the different grammatical dependecies in Tamil sentences and tag them with the appropriate grammatical structure.
Morphological Analysis
Morphological analysis of a word is the process of segmenting the word into component morphemes and assigning the correct morphosyntactic information to it. For a given word, a morphological analyser (MA) will return its root word and the word class along with return its root word and the word class along with the other grammatical information depending upon its word class. Tamil is a verb-final , relatively free word order language which has a rich inflectional and derivational morphology. Therefore building an efficient morphological analyser is challenging. Finite-state automatas and SVM based approaches have been used in the past to build Tamil morphological analysers. This project would aim to build a Tamil Morphological Analyser module that could efficiently retrieve the root form of a given Tamil word from its inflected form and this should be applicable to any domain.
Morphological Generation
The Morphological Generator takes lemma and a Morpho-lexical description as input and gives a word-form as output. It is a reverse process of Morphological Analyzer. The Morphological Generator takes lemma and a Morpho-lexical description as input and gives a word-form as output. It is a reverse process of Morphological Analyzer. Morphological Generator is an essential tool used in Natural Language Processing (NLP) applications for the generation of final fully-inflected words from the root and a set of morphological properties. A morphological generator plays an indispensable role in target language sentence generation in Machine Translation (MT) systems. It is also used in Information Retrieval (IR) systems in the front-end for query expansion. Finite state based morphological generators have proved to work efficiently for Tamil language in the past. In this project we aim to build a Tamil Morphological generator module and this would be built alongside the Morphological analysis module.
Coreference Resolution
Coreference analysis also known as record linkage or identity uncertainty is a difficult and important problem in natural language processing. In many domains we could find multiple views, descriptions or names for the same underlying object. Correctly resolving these references is a necessary task to further processing and understanding the data. This analysis finds the nouns and phrases that refer to the same entity enabling the extraction of relations among entities as well as more complex propositions. Simple pattern matching heuristics have been used in the process of developing a named entity coreference resolver for other languages. Research in this area is still in early stages for Tamil and therefore in this project we aim to study this problem and build an efficient named entity coreference analyser module.
Anaphora Resolution
Anaphora is a relation of interpretational dependency between an antecedent referring expression and an anaphoric referring expression. In general, one considers that the referent of the latter is determined by knowledge inferred from the former. The Tamil pronominals are ‘avan’, ‘aval’ and ‘atu’ are the third person singular pronouns. ‘atu’ is third person singular neuter pronoun. The relation that is established between a pronoun and its antecedent helps to provide more information that can be extracted. Conditional random fields and Decision tree learning have been used before in building an automatic anaphora resolution system for Tamil. In this project we aim to build an Anaphora resolution module that could resolve the anaphora in Tamil text and output referring expressions.
Word sense disambiguation
Word sense disambiguation is the process of identifying the right sense for a word when a word might have two or more meanings. This is vital for problems such as machine translation, question and answering or information retrieval. The need for disambiguating competing senses of ambiguous words is a crucial issue for all the Natural Language Processing activities including machine translation. Support vector machines are efficient in the classification process of discriminating the contexts there by selecting the correct sense of the target word. Unsupervised approaches based on clustering methods have been used as well. In this project we aim to build a module for Tamil word sense disambiguation, in which given Tamil text the correct sense of words needs to be identified.
Named Entity Recognition
Named Entity Recognition is the process of identifying and recognizing named entities such as person, organization, location, date, time and money in the text documents. Named Entity Recognition is a subtask of Information Extraction and Information Extraction is the process of extracting the useful information from data. Unlike English, there is no concept of capital letters in Tamil and hence no capitalization information is available for named entities in Tamil. All named entities are nouns and hence are Noun Phrases but not all Noun Phrases are Named Entities. Conditional random fields and hidden markove models have been used in the past for named entity recognition but it has been done in a very small cale. In this project we aim to build a Named Entity recogniser module for Tamil which could identify and tag named entities in a given text.
Machine Translation and Transliteration
Translation between Tamil and English
Machine Translation (MT) is one of the most important applications of Natural Language Processing, enabling communication and knowledge exchange across linguistic communities. Over the past decades, machine translation research has evolved from rule-based systems and statistical approaches to neural machine translation and large language model-based systems. While significant progress has been achieved for major world languages, there remains a need for high-quality translation systems that accurately capture the linguistic, cultural, and contextual nuances of Tamil.
We aim to develop scalable and accurate machine translation systems for both Tamil-to-English and English-to-Tamil translation. This requires the creation of large-scale parallel corpora, linguistic resources, evaluation datasets, and domain-specific training material covering literature, education, media, government documents, and cultural heritage content. Particular emphasis will be placed on preserving semantic meaning, grammatical correctness, and cultural context during translation.
Beyond conventional translation, this research will support multilingual access to Tamil digital archives, knowledge repositories, historical newspapers, literary collections, and educational resources. By enabling wider access to Tamil knowledge across linguistic boundaries, these systems will contribute to research, digital preservation, public engagement, and cultural exchange.
Translation between Tamil and Sinhala
The development of machine translation systems between Tamil and Sinhala is essential for improving communication and knowledge sharing between the two major linguistic communities of Sri Lanka. Such technologies can help reduce language barriers and enhance access to education, public services, research, and cultural resources.
We aim to develop modern machine translation systems for both Tamil-to-Sinhala and Sinhala-to-Tamil translation. Since both languages possess rich morphological structures and unique linguistic characteristics, translation remains a challenging task that requires extensive language resources, morphological processing tools, and high-quality parallel corpora. The project will focus on creating the necessary linguistic infrastructure and training datasets required for developing accurate translation models.
Research will investigate statistical, neural, and transformer-based machine translation approaches, with particular emphasis on multilingual language models and domain-adaptive systems capable of handling literary, historical, administrative, and conversational content. The resulting systems should be scalable, accurate, and suitable for deployment in applications that serve the wider community.
Translation between Tamil and other Dravidian Languages
The Dravidian language family, which includes Tamil, Malayalam, Telugu, Kannada, and several other languages, shares many linguistic and historical characteristics. These similarities provide valuable opportunities for developing translation systems that facilitate communication, cultural exchange, and collaborative research across the region.
We aim to develop translation systems between Tamil and other Dravidian languages through collaboration with academic institutions, research organizations, and language technology communities. This work will involve the creation of parallel corpora, lexical resources, translation benchmarks, and multilingual language models that support translation in multiple directions.
In addition to direct translation, this research will explore multilingual and cross-lingual approaches that leverage shared linguistic structures among Dravidian languages. These systems will support access to literature, historical records, educational materials, digital archives, and cultural heritage resources while strengthening cooperation between institutions working on Dravidian language technologies.
Machine Transliteration between English and Tamil
Transliteration is the process of converting words from one writing system into another while preserving their pronunciation. It plays a critical role in machine translation, multilingual information retrieval, digital libraries, semantic search, and cross-lingual knowledge access. Accurate transliteration is particularly important for personal names, place names, technical terminology, institutional names, and other named entities that often appear across multiple languages and scripts.
We aim to develop robust transliteration systems between English and Tamil that preserve phonetic accuracy while accounting for the linguistic characteristics of both languages. Transliteration presents several challenges, including differences in phonological systems, multiple possible mappings for vowels and consonants, and variations in pronunciation across dialects and regions. Special attention will be given to handling proper names, historical terminology, and culturally significant concepts that frequently appear in archival and scholarly collections.
Research will investigate statistical, neural, and transformer-based transliteration approaches capable of learning complex character and phoneme mappings. In addition to English-to-Tamil transliteration, we aim to support bidirectional transliteration and extend the framework to Sinhala and other relevant languages. These systems will serve as essential components of machine translation, multilingual search, entity linking, knowledge graph construction, digital archives, and AI-powered knowledge retrieval platforms.
Human-Machine Interface
Tamil Text-to-Speech Generation
Text-to-Speech (TTS) technology enables computers to convert written text into natural and intelligible speech. High-quality speech synthesis plays a critical role in improving accessibility, education, digital inclusion, and human-computer interaction. A modern Tamil text-to-speech system should generate speech that is natural, expressive, and understandable across different listening environments while accurately preserving pronunciation, prosody, and linguistic nuances.
We aim to develop advanced Tamil Text-to-Speech systems capable of generating high-quality speech from Tamil text. This research will investigate statistical, neural, and transformer-based speech synthesis approaches and develop resources required for both Indian Tamil and Sri Lankan Tamil. Particular emphasis will be placed on creating speech technologies that support digital archives, educational platforms, assistive technologies, conversational AI systems, and voice-enabled knowledge access applications.
The resulting systems will form a foundational component of multilingual Voice AI platforms and enable spoken access to Tamil knowledge repositories, cultural heritage collections, and digital learning resources.
Tamil Speech-to-Text Generation
Automatic Speech Recognition (ASR) enables spoken language to be converted into machine-readable text, creating more natural and accessible user interfaces. Speech recognition technologies are increasingly important in voice assistants, digital archives, accessibility tools, search systems, transcription services, and conversational AI applications.
We aim to develop robust Tamil Speech-to-Text systems capable of accurately transcribing spoken Tamil across multiple domains, dialects, and speaking styles. Research will focus on speech corpus development, acoustic modelling, language modelling, neural speech recognition architectures, and multilingual speech processing techniques.
Special attention will be given to Sri Lankan Tamil speech resources and low-resource speech recognition challenges. The resulting systems will support voice-based search, digital content creation, archive transcription, conversational AI, and multilingual communication platforms.
Speaker Recognition Systems for Tamil
Speaker recognition technologies use characteristics of an individual’s voice to identify or verify identity. Such systems have applications in biometric authentication, security, digital services, forensic analysis, and personalized human-computer interaction.
We aim to develop speaker recognition systems capable of identifying and verifying Tamil speakers using advanced machine learning and speech processing techniques. Research will focus on feature extraction, speaker modelling, voice biometrics, and robust recognition under varying acoustic conditions.
These systems may be integrated into future voice-enabled services, digital identity platforms, and research applications involving spoken Tamil.
Speaker Recognition Database for Tamil Speakers
The availability of high-quality speech datasets is critical for the development and evaluation of speech technologies. Reliable speaker recognition databases support research in speech recognition, speaker identification, speech synthesis, and voice biometrics.
We aim to create a comprehensive speech database representing Tamil speakers across different regions, age groups, genders, dialects, and speaking styles. Particular attention will be given to capturing both Indian Tamil and Sri Lankan Tamil speech varieties. The resulting resource will serve as an open research dataset supporting the development and benchmarking of future speech technologies.
Deep Learning Approaches for Speech Recognition
Recent advances in deep learning have transformed speech recognition by significantly improving accuracy and robustness across a wide range of applications. Neural architectures such as Deep Neural Networks (DNNs), Recurrent Neural Networks (RNNs), Transformer models, and self-supervised learning approaches have become the dominant technologies in modern speech processing.
We aim to investigate and develop advanced deep learning approaches for Tamil speech recognition. Research will focus on improving recognition accuracy, handling dialectal variation, reducing data requirements, and developing models suitable for low-resource language environments. These technologies will contribute to the development of next-generation speech recognition systems and multilingual Voice AI applications.
Modern Input Methods for Tamil
The growing use of mobile devices, touch interfaces, and emerging interaction technologies creates a need for efficient and intuitive Tamil input methods. Traditional keyboard-based approaches are often insufficient for modern digital environments, particularly on mobile and wearable devices.
We aim to develop modern Tamil input systems that support touchscreen, handwriting, gesture-based, voice-based, and predictive text entry. Research will focus on intelligent text prediction, language modelling, adaptive keyboards, multilingual input systems, and accessibility-oriented design.
These technologies will improve the usability of Tamil in digital environments and contribute to wider adoption of Tamil computing across mobile devices, educational platforms, communication systems, and emerging human-computer interaction technologies.
Voice AI and Conversational Interfaces
Advances in speech technologies, large language models, and conversational AI have created new opportunities for voice-based interaction with digital knowledge systems. By combining speech recognition, natural language understanding, knowledge retrieval, and speech synthesis, users can interact with information repositories using natural spoken language.
We aim to develop multilingual Voice AI systems that enable speech-based search, conversational assistance, knowledge discovery, and educational interaction in Tamil. These systems will integrate speech recognition, text-to-speech generation, language models, and knowledge retrieval technologies to provide natural access to digital archives, research collections, and cultural heritage resources.
This research will support the development of voice-enabled knowledge platforms that make Tamil information more accessible, discoverable, and usable for diverse communities worldwide.
Evolutionary Study of Language
Evolution of Tamil
Evidences from archaeology and literature demonstrate that Tamil is one of the ancient languages in the world. Geological, cultural, and synthetic influences made the language to evolve in the past. Besides preserving the language, it is important to understand the process of evolution of the language to develop an in depth understanding of the current form of Tamil. This novel research studies about developing a mathematical model for representing the trends of the Tamil Language. Opportunities of exploiting cutting edge parallel computing and simulation techniques are considered in this task. The fundamental research in this task includes finding interconnections with other language proximities, such as Sanskrit and comparing present language forms of the geographical regions where Tamil believed to be the root language.
Supporting old Tamil scripts
Current Tamil script has evolved through many centuries. Important paleographic information are preserved in paleographic scripts. This area focusses on creating appropriate fonts that enable the faithful digitization and representation of these paleographic documents for research and disseminations. We would also focus on encoding these old scripts and any rare symbols into the Unicode standard.
Digital Paleography for Tamil
Digital Paleography is an area that is concerned with the dating of texts and writer recognitions based on the scribal handwriting present in paleographic texts. Tamil being on the earliest attested languages in India, there are thousands of paleographic artefacts such as epigraphs that span across two millennia. In many case, it is very hard to date these document to ascertain their provenance. In such cases, Digital Paleographic methods can be extremely useful. Similar to OCR, the characters are converted into feature vectors and are compared against a set of known characters, to place a text under a certain period of time or scribe.
Supporting Tamil philology
Philology is the study of literary texts. Frequently, they are involved in ascertaining the authenticity of texts, studying text variations and producing critical edition of texts that are based on multiple sources of the text. Bulk of this work is currently done manually, this project helps to introducing computational techniques to enable Tamil philologists to produce critical texts more effectively.
Knowledge Engineering
Knowledge Retrieval
The Noolaham AI ecosystem aims to transform large-scale Sri Lankan Tamil digital archives into a structured and semantically enriched knowledge base. While traditional digital libraries primarily support document and keyword-based retrieval, the proposed knowledge engineering framework enables higher levels of information access through semantic search, entity exploration, and knowledge retrieval. By combining corpus creation, NLP enrichment, taxonomy and ontology development, and knowledge graph construction, the system converts unstructured textual content into interconnected knowledge representations.
Knowledge retrieval allows users to discover relationships, events, concepts, people, organizations, and places across thousands of archival documents without relying solely on exact keyword matches. Ontology-aligned knowledge graphs provide the foundation for semantic reasoning and contextual information access, enabling users to navigate and query historical and cultural knowledge more effectively.
Applied to the Noolaham digital archives, these capabilities can support advanced research applications such as historical timeline generation, entity relationship discovery, trend analysis, and exploration of cultural, social, and political developments reflected in Sri Lankan Tamil publications over time. Through AI-powered retrieval and reasoning, the ecosystem facilitates deeper understanding and new insights from large-scale archival collections that would otherwise remain difficult to analyze manually.