Nauman Dawalatabad
Research Scientist at Zoom
Sunnyvale, California, United States
Summary
🤩
Rockstar🎓
Top SchoolNauman Dawalatabad is a research scientist with a decade of experience building scalable Speech AI and multilingual ASR systems, currently advancing conversational AI at Zoom after a postdoc at MIT CSAIL. He holds a PhD from IIT Madras and has bridged academia and industry through roles at Samsung, Mila, and research projects sponsored by DRDO, focusing on on-device speech recognition, speaker recognition, and diarization. An active open-source contributor to the prominent SpeechBrain toolkit, he implemented Xvector recipes and practical examples that helped operationalize speaker embedding workflows. Based in Sunnyvale, he combines deep research rigor with hands-on engineering to move models from prototype to production, and his background training students and shipping on-device systems gives him a rare blend of mentorship and productization experience.
10 years of coding experience
7 years of employment as a software developer
Indian Institute of Technology MadrasVisiting Research Student, Visiting Research Student at Mila - Quebec Artificial Intelligence InstituteBachelor of Engineering - BE Information Technology, Bachelor of Engineering - BE Information Technology at University of Mumbai
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (4)
speaker-diarization10
pytorch10
speaker-recognition9
machine-learning8
Programming languages (1)
Github contributions (5)
speechbrain/speechbrain
Mar 2020 - Aug 2022
A PyTorch-based Speech Toolkit
Role in this project:
ML Engineer
ML EngineerContributions:4 reviews, 385 commits, 2 PRs in 2 years 5 months
Contributions summary:Nauman's contributions primarily involved the addition and refinement of the "Standard Xvector" and related modules, showcasing a focus on speaker recognition and diarization. The commit messages and code changes demonstrate the implementation of a crucial feature in the speech toolkit, as well as the development of an example experiment for the Xvector model and modifications to improve documentation. The user's work includes adding a minimal example for Xvector and the addition of a full recipe for Xvector training within the VoxCeleb framework, solidifying their role in model implementation and practical application.
voice-recognitionasrspeech-recognitionspeech-separationspeaker-verification
nauman-daw/speechbrain-1
Sep 2021 - Aug 2022
A PyTorch-based Speech Toolkit
Contributions:2 pushes, 1 branch in 11 months
pytorchspeech-toolkitasrspeech-recognitionspeech-synthesis
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.