Aims

Develop a suite of language resources, including parallel corpora for machine translation between Indian languages and benchmark datasets for key NLP tasks like Part-of-Speech (PoS) tagging, Named Entity Recognition (NER), Automatic Speech Recognition (ASR), and Text-to-Speech (TTS).

Primary Focus

50% domain-specific content from a mix of sectors: science, healthcare, agriculture, climate, tourism, and judiciary. 30% educational content, focusing on academic materials, textbooks, and learning resources. 20% conversational and governance content, including social media, dialogues, and official documents.

Key Focus

Standardize, preserve, and create language resources to support NLP research and applications.

Applications

This dataset will provide translation between Tamil and Kannada, Malayalam, Telugu and it will also enable the development of multilingual chatbots. This will facilitate more effective cross-sector communication.

Intiative

Establish a centralized repository for Indian language data and provide a platform for developing and benchmarking Human Language Technology (HLT) applications.

Results

To ensure rapid and efficient translation, the final output requires minimal post-editing, making it immediately suitable for publication on print and digital platforms.

Development

Create comprehensive leaderboards to systematically evaluate the performance of models in key Natural Language Processing (NLP) tasks. These leaderboards will serve as benchmarks for assessing machine translation (MT), Part-of-Speech (PoS) tagging, Named Entity Recognition (NER), Natural Language Generation (NLG), sentiment analysis, Automatic Speech Recognition (ASR), and Text-to-Speech (TTS).

Delivery

These advanced, no-cost tools offer a superior alternative to existing solutions like Google Translate, while simultaneously contributing to the preservation and growth of our linguistic heritage.

Step 1: Identification of Tamil language resources

Identify and list existing Tamil datasets and tools to understand the current resource availability.

Step 2: Acquisition of resources across target domains

To ensure comprehensive coverage of real-world language usage, data will be collected from key domains, including Education, Governance & Policy, Judiciary, Science & Technology, Healthcare, Agriculture, Climate, and Tourism.

Step 3: Creation of language resources and benchmarks

Develop language resources and datasets for Machine Translation and NLP, and establish benchmarks to systematically evaluate tool performance.

Step 4: Development of MT evaluation leaderboards

Develop MT evaluation leaderboards to assess translation systems, enabling performance tracking and fostering continuous improvement.

Step 5: Leaderboards for ASR and TTS technologies

Develop evaluation scoreboards for speech recognition and synthesis systems to measure accuracy, clarity, and overall performance.

Step 6: Benchmarks for PoS and NER taggers

Define evaluation protocols for tools such as PoS and NER taggers to ensure fair and consistent assessment.

Commander(Dr) Anil Rana

Director

Manipal Institute of Technology

MAHE, Manipal, India

Dr. Chandrakala C B

Joint Director

Additional Professor

School of Computer Engineering

MIT, MAHE Manipal, India

Dr. Radhika M Pai

Dean and Professor

School of Computer Engineering

MIT, MAHE Manipal, India

Dr. Muralikrishna SN

Principal Investigator

Associate Professor

School of Computer Engineering

MIT, MAHE Manipal, India

Dr. Ashalatha Nayak

Co-Investigator

Professor

School of Computer Engineering

MIT, MAHE Manipal, India

Dr. Ashwath Rao B

Co-Investigator

Assistant Professor - Selection Grade

School of Computer Engineering

MIT, MAHE Manipal, India

Dr. Raghavendra Ganiga

Co-Investigator

Associate Professor

School of Computer Engineering

MIT, MAHE Manipal, India

Mr. Ganesh Babu C

Co-Investigator

Assistant Professor Senior Scale

School of Computer Engineering

MIT, MAHE Manipal, India

Dr. Raghurama Holla

Co-Investigator

Assistant Professor

School of Computer Engineering

MIT, MAHE Manipal, India

Freelance Translators and Reviewer

We are looking for Freelance Translators and Reviewer proficient in Dravidian languages. The language pair for translation are:

• Tamil - Kannada
• Tamil - Malayalam
• Tamil-Telugu

Note that Tamil is the source language

• Proficiency in Tamil
• Proficient in either Kannada, Malayalam or Telugu
• Basic Computer Knowledge