Nauman Dawalatabad is a postdoctoral researcher at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), USA, under the supervision of Dr. James Glass in the Spoken Language Systems (SLS) group. Prior to joining MIT, he was a Lead Engineer at Samsung Research, Bangalore, India, where he worked on on-device speech recognition. He obtained his Ph.D. (with the Institute Research Award) in Computer Science and Engineering from the Indian Institute of Technology Madras (IIT Madras), India, working under the supervision of Prof. C. Chandra Sekhar and Prof. Hema A. Murthy. During his PhD, he was also a visiting research student and a core team member of the SpeechBrain group at Mila - Quebec AI Institute, Montreal, Canada, under supervision of Prof. Yoshua Bengio and Prof. Mirco Ravanelli. In 2023, he is selected as one of the IEEE ICASSP Rising Stars in Signal Processing. He serves as a reviewer/PC member for various Speech/NLP conferences and journals.

Research interests: Robust speech recognition, speaker recognition, speaker diarization, voice conversion, audio deepfakes, speech in healthcare, on-device ASR, and multimodal conversational AI.

Reviewer/PC member:  ICASSP, Interspeech, IEEE/ACM T-ASLP, IEEE SPL, IEEE T-PAMI,  Speech Communication, WASPAA, ACL, EMNLP, ICML SPIGM, and others

Connect on:  LinkedIn  |  X (Twitter)  |  Google Scholar  |  GitHub

Updates/News:

• Featured in MIT News and MIT CSAIL website, 2024.
• A paper accepted at IEEE ICASSP 2024 Self-supervision in Audio, Speech and Beyond Workshop. (Paper: Cross-Lingual Transfer Learning for Speech Translation)
• Attending SANE 2023 workshop at NYU.
• Invited alumni guest talk (Alum@Alma) at IIT Madras, 2023. (Host: Prof. Rupesh Nasre and Prof. N.S. Narayanaswamy)  
• Selected as IEEE ICASSP Rising Stars in Signal Processing 2023.
• A paper accepted at IEEE ICASSP 2023. (Paper: Unsupervised Uncertainty based Speech Data Pseudo-label Filtering and Model Calibration)
• Invited guest lecture at University of Groningen in Speech Recognition II course, Netherlands, 2023. (Host: Prof. Shekhar Nayak)
• A poster accepted at MIT-LIDS workshop, MIT, 2023.
• Invited keynote speaker at RAMMML 2023.
• A paper accepted at EMNLP, ACL Findings, 2022. (Paper: Speaker Diarization+DementiaDetection+ASR decoding for Long Interviews)
• Invited guest lecture at MIT EECS department for the MIT-6.345 Speech Recognition Course, 2022. (Host: Prof. James Glass )
• Our work on model compression for ondevice ASR featured in Samsung blog post, 2022.
• A paper accepted at Interspeech 2022. (Paper: Conformer-Transducer Model Compression for On-device ASR)
• Samsung Research E-Spot award 2021. 
• Two papers accepted at IEEE ASRU 2021. (Paper 1: ASR RNN-Transducer On-device Model Compression  ; Paper 2: E2E ASR with HiTNet )
• Invited guest talk at Avignon University, France, 2021. (Host: Prof. Titouan Parcollet)

Office Address:
MIT Computer Science and Artificial Intelligence Laboratory,
Massachusetts Institute of Technology,
32 Vassar Street, 32-G442,
Cambridge, MA 02139.