Akhilesh Kakolu Ramarao

Twitter LinkedIn Github

Hello! I am Akhilesh Kakolu Ramarao. I am a Natural Language Processing Researcher, currently working as a Wissenschaftlicher Mitarbeiter at the Department of English Language and Linguistics, Heinrich Heine University, Germany.

I currently work on Computational Morphology – supervised by Prof. Dr. Kevin Tang and Dr. Dinah Baer-Henney.

In my previous life, I have worked on building multilingual chatbots and voice assistants at Skit.ai, as an NLP Researcher at CivicDataLab and as a Software Engineer at Ramco Systems.

I am actively engaged in language revitalization efforts for indigenous communities in Arunachal Pradesh, India, with a focus on Idu Mishmi and K’man Mishmi languages.

Research Interests

  • Computational Morphology
  • Low-resource NLP

Skills

  • Recent experience: Python, Bash, LaTeX, Scheme, Julia

  • Past experience: Haskell, Dart, JavaScript (AngularJS)

  • Frameworks: PyTorch, Flutter

  • Platforms: Arch linux, Ubuntu

  • Cloud: AWS, GCP, High-performance computing (HPC)

Academic Positions

Heinrich-Heine-Universität Düsseldorf, Germany

2021-present

Position: Wissenschaftlicher Mitarbeiter, Department of English Language and Linguistics

  • Teaching computational linguistics courses for BA and MA students. Course list
  • Responsible for IT administration for Slamlab

Education

M.A., Linguistics

Heinrich-Heine-Universität Düsseldorf, Germany, 2022-2024

B.E., Automobile Engineering

Dayananda Sagar College of Engineering, Bengaluru, India, 2012-2016

Teaching

Publications

  • Kakolu Ramarao, A., Petersen, W., Stein, A., Stein, E., & Xia, H. (2024). KlarTextCoders at StaGE: Automatic Statement Annotations for German Easy Language. In Proceedings of GermEval 2024 Shared Task on Statement Segmentation in German Easy Language (StaGE) (pp. 15–27), Vienna, Austria. Association for Computational Lingustics. https://aclanthology.org/2024.germeval-1.2.pdf

  • Muschalik, J., Schmitz, D., Kakolu Ramarao, A. et al. Typing /s/-morphology between the keys?. Read Writ (2024). DOI

  • Jeong, C., Schmitz, D., Kakolu Ramarao, A., Stein, A., & Tang, K. (2023). Linear discriminative learning: a competitive non-neural baseline for morphological inflection. In Proceedings of the 20th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology (pp. 138-150). Association for Computational Linguistics. https:// aclanthology.org/2023.sigmorphon-1.16.pdf

  • Kakolu Ramarao, A., Zinova, Y., Tang, K., & van de Vijver, R. (2022). HeiMorph at SIGMORPHON 2022 shared task on morphological acquisition trajectories. In Proceedings of the 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology (pp. 236-239). Association for Computational Linguistics. https:// aclanthology.org/2022.sigmorphon-1.24

Talks

Peer-reviewed Talks

  • “Community-driven Language Revitalization: Translating Children’s Literature in K’man Mishmi” at International Conference on Language, Ecology and Culture 2024, NEILAC, Guwahati, India. Team: Akhilesh Kakolu Ramarao, Satyanarayan Mundayoor, Pamir Gogoi, Sakelu Chikro. Presenter: Sakelu Chikro. PDF

  • “KlarTextCoders at StaGE: Automatic Statement Annotations for German Easy Language” at KONVENS 2024, University of Vienna, Austria. Team: Akhilesh Kakolu Ramarao, Wiebke Petersen, Anna Sophia Stein, Emma Stein, Hanxin Xia. Presenter: Akhilesh Kakolu Ramarao and Emma Stein. PDF

  • “A computational approach to understanding the cognitive reality of morphomic patterns: the case of L-shaped morphomes in Spanish” at Annual Meeting of Linguistics Association of Great Britain and Northern Ireland 2024, Newcastle University, United Kingdom. Team: Akhilesh Kakolu Ramarao, Kevin Tang and Dinah Baer-Henney. Presenter: Akhilesh Kakolu Ramarao PDF

  • “Acoustic Analysis of Vowels in Idu Mishmi” at the 27th Himalayan Languages Symposium, 2024, IIT Guwahati, India. Abstract. Presenter: Pamir Gogoi. Team: Pamir Gogoi, Akhilesh Kakolu Ramarao and Ria Borah Sonowal.

  • “From Zero to Terminal Hero” at the 32nd Tagung der Computerlinguistik-Studierenden (TaCoS), 1 July 2023, Heinrich Heine University, Düsseldorf, Germany. PDF

  • “Can Neural Networks learn human-like behavior when inflecting verbs? The case of Spanish” at the Five-Minute Linguist competition, Linguistic Society of America 97th Annual Meeting, 2023

  • “Modeling irregular morphological patterns using Transformers” at HeiCAD Lightning Talks, 2022

Invited Talks

  • Language Revitalization: A case of Idu Mishmi at Microsoft Research India, 2021
  • Language Revitalization at Navana Tech, 2021

Poster Presentations

Peer-reviewed Posters

  • Ramarao, A. K., & Stein, A. S. AnglistikVoices: an L2 English speech dataset for educational and technological advancement in speech technology. LabPhon 19, Hangyang University, Seoul, South Korea, 2024. PDF

  • Muschalik, J., Schmitz, D., Kakolu Ramarao, A., & Baer-Henney, D. (2023). Acoustic duration and typing timing - same, same… but different? Phonetics and Phonology in Europe 2023, Nijmegen, Netherlands.

  • Tang, K., Kakolu Ramarao, A., & Baer-Henney, D. (2022). Modeling irregular morphological patterns with recurrent neural network: the case of the L-shaped morphome. 13th Mediterranean Morphology Meeting, University of the Aegean, Greece.

Invited Posters

  • Kakolu Ramarao, A. (2020). Analysis of POCSO cases in Uttar Pradesh. Agami, India, Bengaluru.

Demos

Grants/Awards

  • First prize at the HHU Legal Hackathon 2022. Team: Akhilesh Kakolu Ramarao, Phaedon Paschalis and Diana Schill.
  • First prize at the HeiCAD Lightning Talks 2022.
  • Hardware Grant. 20 × NVIDIA® JetsonTM Nano Developer Kits. NVIDIA Academic Hardware Grant, “FOSStering Digital Humanities through Accent Diversity & Conversational Devices”. Co-applicant: Kevin Tang
  • Education Grant. €9,424. e-Learning support fund, Heinrich-HeineUniversity Düsseldorf. “Interactive web-based review units for phonetics and phonology”. Co-applicant: Christopher Geissler and Kevin Tang

Conference/Workshop Organization

  • Translation Workshop: English to K’man Mishmi, 2023, Bamboosa Library, Tezu, Arunachal Pradesh, India
  • Chief Organizer, LibreSlam - Free and Open Source (FOSS) for everybody, Heinrich-Heine-University, Germany, 2022
  • Chief Organizer and Teaching Member, LY-CS101 2020, National College, Bengaluru, India.

Industry Experience

CivicDataLab

2019-2020

Position: NLP Researcher

  • Developed Language Platform for Assamese, including Optical Character Recognition, Machine Translation system, keyboard, rich text editor, and parallel corpora creation
  • Conducted case studies on language complexity of Union Budget Speeches and topic modeling on Union Budget suggestions
  • Built crowd-sourced translation platform for COVID-19 information dissemination in local languages
  • Involved in developing a text annotation platform for sequence labeling of entities and relations in Indian court judgments

Skit.ai (previously, Vernacular.ai)

2017-2019

Position: Software Engineer

  • Involved in building multilingual chatbots and voice assistants.
  • Implemented conversation design framework, dialog management system, and user response prediction strategies

Ramco Systems

2016-2017

Position: Software Engineer

  • Developed a chatbot for Human Resource (HR) system, responsible for bug fixes, documentation, and feature implementation of machine learning components

Volunteering

Idu Mishmi Cultural and Literary Society (IMCLS)

2020-present

Role: Chief Coordinator

  • Managing partnerships and technology
  • Developed an android E-Dictionary application for the Idu Mishmi language

Lohit Youth Library Network

2020-present

Role: Volunteer

Consulting

XRI Institute

2020-2021

Vrook

2020-2021

Internships

Sense Infinity Technologies

Position: Full-stack developer intern

Maruti Suzuki

Position: Industry trainee