About Me

Hi! I’m Devang, an Applied Scientist II at Amazon AWS in Toronto, Canada, specializing in Automatic Speech Recognition (ASR). My work involves addressing challenges in multilingual support, continual learning, Audio Language Model (LLM) integration, low-resource scenarios, and domain-specific personalization. I also contributed to the HealthScribe project at Amazon, focusing on abstractive summarization using Language Models (LLMs).

As an M.Sc. student at the MILA Lab and McGill University, under the Prof. Siva Reddy and as a research intern at Korbit.AI, our collaborative efforts centered on neural question generation from open educational resources like Wikipedia, specifically in the statistics and Machine Learning domains.

Experience

OrganizationPositionDuration
Amazon CanadaApplied ScientistJuly 2022 - Present
Amazon UKApplied Science InternJul 2021 - Oct 2021
Korbit.AIResearch Sciene InternSep 2020 - May 2021
Amazon IndiaSoftware EngineerSep 2018 - Aug 2020
INRIA Labs FranceResearch InternMay 2018 - Aug 2018
Amazon IndiaSoftware Engineer InternMay 2017 - Jul 2017
Busigence IndiaData Science InternDec 2016
CFILT Lab IIT BombayR&D InternMay 2016 - Jul 2016

Publications

In chronological order:

  1. SM. Jayanthi, D. Kulshreshtha, S. Dingliwal, S. Ronanki, S. Bodapati: Retrieve and Copy: Scaling ASR Personalization to Large Catalogs. EMNLP 2023
  2. VR. Elluru, D. Kulshreshtha, R. Paturi, S. Bodapati, S. Ronanki: Generalized zero-shot audio-to-intent classification. ASRU 2023
  3. D. Kulshreshtha, S. Dingliwal, B. Houston, S. Bodapati: Multilingual contextual adapters to improve custom word recognition in low-resource languages. INTERSPEECH 2023
  4. N. Das, M. Sunkara, S. Bodapati, J. Cai, D. Kulshreshtha, J. Farris, K. Kirchhoff: Mask the Bias: Improving Domain-Adaptive Generalization of CTC-Based ASR with Internal Language Model Estimation. ICASSP 2023
  5. D. Kulshreshtha, M. Shayan, R. Belfer, S. Reddy, IV. Serban, E. Kochmar: Few-shot question generation for personalized feedback in intelligent tutoring systems. IJCAI-PAIS 2022
  6. D. Kulshreshtha, R. Belfer, I.V. Serban, S. Reddy: Back-Training excels Self-Training at Unsupervised Domain Adaptation of Question Generation and Passage Retrieval. EMNLP 2021
  7. D. Kulshreshtha, P. Goel, A.K. Singh: How emotional are you? Neural Architectures for Emotion Intensity Prediction in Microblogs. COLING 2018
  8. H. Rangwani, D. Kulshreshtha, A.K. Singh: NLPRL-IITBHU at SemEval-2018 Task 3: Combining linguistic features and emoji pre-trained CNN for irony detection in tweets. SemEval at NAACL 2018
  9. P. Goel, D.Kulshreshtha, P. Jain, K.K. Shukla: Prayas at EmoInt 2017: An Ensemble of Deep Neural Architectures for Emotion Intensity Prediction in Tweets. WASSA at EMNLP 2017
  10. D. Kulshreshtha: Feature Augmented Deep Neural Networks for Collaborative Filtering. IJCAI 2017 Workshop on AI Applications in e-commerce