Publications

You can also browse my Google Scholar profile.

2024

• Faster Speech-LLaMA inference with multi-token prediction
  Desh Raj, Gil Keren, Junteng Jia, Jay Mahadeokar, Ozlem Kalinli
  Submitted to IEEE ICASSP 2025
  Paper

• M-BEST-RQ: A multi-channel speech foundation model for smart glasses
  Yufeng Yang, Desh Raj, Ju Lin, Niko Moritz, Junteng Jia, Gil Keren, Egor Lakomkin, Yiteng Huang, Jacob Donley, Jay Mahadeokar, Ozlem Kalinli
  Submitted to IEEE ICASSP 2025
  Paper

• ConEC: Earnings Call Dataset with Real-world Contexts for Benchmarking Contextual Speech Recognition
  Ruizhe Huang, Mahsa Yarmohammadi, Jan Trmal, Jing Liu, Desh Raj, Leibny Paola Garcia, Alexei V Ivanov, Patrick Ehlen, Mingzhi Yu, Dan Povey, Sanjeev Khudanpur
  LREC 2024
  Paper

• Listening to multi-talker conversations: Modular and end-to-end perspectives
  Desh Raj
  PhD Thesis, Johns Hopkins University
  Thesis Slides Video

• On speaker attribution with SURT
  Desh Raj, Matthew Wiesner, Matthew Maciejewski, Paola Garcia, Daniel Povey, Sanjeev Khudanpur
  Speaker Odyssey 2024
  Paper Slides

• Updated corpora and benchmarks for long-form speech recognition
  Jennifer Drexler Fox, Desh Raj, Natalie Delworth, Quinn McNamara, Corey Miller, Migüel Jetté
  IEEE ICASSP 2024
  Paper Code

• Training Early-Exit Architectures for Automatic Speech Recognition: Fine-Tuning Pre-Trained Models or Training from Scratch
  George August Wright, Umberto Cappellazzo, Salah Zaiem, Desh Raj, Lucas Ondel Yang, Daniele Falavigna, Alessio Brutti
  IEEE ICASSP 2024 Workshop on Self-supervision in Audio, Speech, and Beyond (SASB)
  Paper

2023

• Learning from flawed data: Weakly supervised automatic speech recognition
  Dongji Gao, Hainan Xu, Desh Raj, Leibny Paola Garcia Perera, Daniel Povey, Sanjeev Khudanpur
  IEEE ASRU 2023
  Paper Code

• SURT 2.0: Advances in transducer-based multi-talker speech recognition
  Desh Raj, Daniel Povey, Sanjeev Khudanpur
  IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASLP)
  Paper ArXiv Poster Webpage

• The CHiME-7 DASR challenge: Distant meeting transcription with multiple devices in diverse scenarios
  Samuele Cornell, Matthew Wiesner, Shinji Watanabe, Desh Raj, Xuankai Chang, Paola Garcia, Matthew Maciejewski, Yoshiki Masuyama, Zhong-Qiu Wang, Stefano Squartini, Sanjeev Khudanpur
  CHiME Workshop at InterSpeech 2023
  Paper Website

• GPU-accelerated guided source separation for meeting transcription
  Desh Raj, Daniel Povey, Sanjeev Khudanpur
  InterSpeech 2023
  Paper ArXiv Poster Code

• Anchored speech recognition using neural transducers
  Desh Raj, Junteng Jia, Jay Mahadeokar, Chunyang Wu, Niko Moritz, Xiaohui Zhang, Ozlem Kalinli
  IEEE ICASSP 2023
  Paper Slides Video

• Adapting self-supervised models to multi-talker speech recognition using speaker embeddings
  Zili Huang, Desh Raj, Paola Garcia, Sanjeev Khudanpur
  IEEE ICASSP 2023
  Paper Code

2022

• Low-Latency speech separation guided diarization for telephone conversations
  Giovanni Morrone, Samuele Cornell, Desh Raj, Luca Serafini, Enrico Zovato, Alessio Brutti, Stefano Squartini
  IEEE Spoken Language Technology (SLT) Workshop 2022
  Paper

• Continuous streaming multi-talker ASR with dual-path transducers
  Desh Raj, Liang Lu, Zhuo Chen, Yashesh Gaur, Jinyu Li
  IEEE ICASSP 2022
  Paper Slides Poster Video

• Injecting text and cross-lingual supervision in few-shot learning from self-supervised models
  Matthew Wiesner, Desh Raj, Sanjeev Khudanpur
  IEEE ICASSP 2022
  Paper Code Poster Video (Matthew)

2021

• Joint speaker diarization and speech recognition based on region proposal networks
  Zili Huang, Marc Delcroix, Leibny Paola Garcia, Shinji Watanabe, Desh Raj, Sanjeev Khudanpur
  Computer, Speech, and Language, Vol. 72
  Paper

• Reformulating DOVER-Lap label mapping as a graph partitioning problem
  Desh Raj, Sanjeev Khudanpur
  INTERSPEECH 2021
  Paper Code Report Slides Video

• Auxiliary loss function for target speech extraction and recognition with weak supervision based on speaker characteristics
  Katerina Zmolikova, Marc Delcroix, Desh Raj, Shinji Watanabe, Jan Černocký
  INTERSPEECH 2021
  Paper

• Target-speaker voice activity detection with improved i-vector estimation for unknown number of speaker
  Mao-Kui He, Desh Raj, Zili Huang, Jun Du, Zhuo Chen, Shinji Watanabe
  INTERSPEECH 2021
  Paper

• Training hybrid models on noisy transliterated transcripts for code-switched speech recognition
  Matthew Wiesner, Mousmita Sarma, Ashish Arora, Desh Raj, Dongji Gao, Ruizhe Huang, Supreet Preet, Moris Johnson, Zikra Iqbal, Nagendra Goel, Jan Trmal, Leibny Garcıa-Perera, Sanjeev Khudanpur
  INTERSPEECH 2021
  Paper Code

• The Hitachi-JHU DIHARD III system: Competitive end-to-end neural diarization and x-vector clustering systems combined by DOVER-Lap
  Shota Horiguchi, Nelson Yalta, Paola Garcia, Yuki Takashima, Yawen Xue, Desh Raj, Zili Huang, Yusuke Fujita, Shinji Watanabe, Sanjeev Khudanpur
  Third DIHARD Speech Diarization Challenge
  Paper

• Multi-class spectral clustering with overlaps for speaker diarization
  Desh Raj, Zili Huang, Sanjeev Khudanpur
  IEEE Spoken Language Technology (SLT) Workshop 2021
  Paper Code Slides

• DOVER-Lap: A method for combining overlap-aware diarization outputs
  Desh Raj, Paola Garcia, Zili Huang, Shinji Watanabe, Daniel Povey, Andreas Stolcke, Sanjeev Khudanpur
  IEEE Spoken Language Technology (SLT) Workshop 2021
  Paper Code Slides

• Integration of speech separation, diarization, and recognition for multi-speaker meetings: System description, comparison, and analysis
  Desh Raj, Pavel Denisov, Zhuo Chen, Hakan Erdogan, Zili Huang, Maokui He, Shinji Watanabe, Jun Du, Takuya Yoshioka, Yi Luo, Naoyuki Kanda, Jinyu Li, Scott Wisdom, John R. Hershey
  IEEE Spoken Language Technology (SLT) Workshop 2021
  Paper Code Slides

• Sequential multi-frame neural beamforming for speech separation and enhancement
  Zhong-Qiu Wang, Hakan Erdogan, Scott Wisdom, Kevin Wilson, Desh Raj, Shinji Watanabe, Zhuo Chen, John R. Hershey
  IEEE Spoken Language Technology (SLT) Workshop 2021
  Paper

2020

• Frustratingly easy noise-aware training of acoustic models
  Desh Raj, Jesus Villalba, Daniel Povey, Sanjeev Khudanpur
  ArXiv, 2020
  Paper Code

• The JHU multi-microphone multi-speaker ASR system for the CHiME-6 challenge
  Ashish Arora*, Desh Raj*, Aswin Shanmugam Subramanian*, Ke Li*, Bar Benyair, Matthew Maciejewski, Piotr Zelasko, Paola Garcia, Shinji Watanabe, Sanjeev Khudanpur.
  The 6th CHiME Workshop (at ICASSP 2020).
  Paper Video Slides

2019

• Probing the infomation encoded in x-vectors
  Desh Raj, David Snyder, Daniel Povey, Sanjeev Khudanpur.
  IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) 2019.
  Paper Code Poster

• Using ASR methods for OCR
  Ashish Arora, Chun Chieh Chang, Babak Rekabdar, Daniel Povey, David Etter, Desh Raj, Hossein Hadian, Jan Trmal, Paola Garcia, Shinji Watanabe, Vimal Manohar, Yiwen Shao, Sanjeev Khudanpur.
  International Conference on Document Analysis and Recognition (ICDAR) 2019.
  Preprint Paper Code Blog

2018

• Uncertain fuzzy self-organization based clustering: interval type-2 approach to adaptive resonance theory
  Shakaiba Majheed, Aditya Gupta, Desh Raj, Frank Chung-hoon Rhee.
  Information Sciences, 2018.
  Paper

2017

• Learning local and global contexts using a convolutional recurrent neural network for relation classification in biomedical text
  Desh Raj, Sunil Kumar Sahu, Ashish Anand.
  Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL) 2017.
  Paper Poster Code

• Analysis of data generated from multidimensional type-1 and type-2 fuzzy membership functions
  Desh Raj, Aditya Gupta, Bhuvnesh Garg, Kenil Tanna, Frank Chung-hoon Rhee.
  IEEE Transactions on Fuzzy Systems, 2017.
  Paper