Muhammad Shakeel
Scholar

Muhammad Shakeel

Google Scholar ID: FmgluMoAAAAJ
Honda Research Institute Japan Co., Ltd.
Speech ProcessingSpeech RecognitionArtificial Intelligence
Citations & Impact
All-time
Citations
449
 
H-index
10
 
i10-index
10
 
Publications
20
 
Co-authors
9
list available
Resume (English only)
Academic Achievements
  • ISCA Best Student Paper Award (INTERSPEECH 2025).
  • IEEE SLT Best Paper Award (2024).
  • Fully funded Ph.D. Research Fellowship from MEXT (2019).
  • Tohoku University Exchange Programme (2013).
  • Erasmus Mundus Scholarship (2011).
  • First-authored paper 'UME' accepted for publication.
Research Experience
  • Currently collaborating with Prof. Shinji Watanabe at Honda Research Institute Japan on end-to-end speech recognition.
  • Contributed to the Open Whisper-style Speech Model (OWSM) project with Carnegie Mellon University’s Language Technologies Institute.
  • Enhanced OWSM with E-Branchformer and developed non-autoregressive OWSM-CTC for improved speed and robustness.
  • Pioneered contextual ASR techniques including dynamic vocabularies, intermediate biasing losses, and bias phrase boosted (BPB) beam search.
  • Co-developed the unified 4D ASR model integrating CTC, Attention, RNN-T, and Mask-CTC decoders.
  • Worked on jointly optimizing streaming and non-streaming ASR and model compression techniques (e.g., joint distillation and pruning in DPHuBERT).