Contact | Education | Research and publications | Thesis | Professional | Teaching | Other skills | PDFPDF 

Research and publications


  1. Efficient Cascaded Streaming ASR System via Frame Rate Reduction
  2. Partial Rewriting for Multi-Stage ASR
  3. Sharing Low Rank Conformer Weights for Tiny Always-On Ambient Speech Recognition Models
  4. Flickering reduction with partial hypothesis reranking for streaming ASR
  5. Neural-FST Class Language Model for End-to-End Speech Recognition
  6. Improved pronunciation prediction accuracy using morphology
  7. Improved Neural Language Model Fusion for Streaming Recurrent Neural Network Transducer
  8. Anti-aliasing regularization in stacking layers
  9. Algorithmic Exploration of American English Dialects
  10. A Streaming On-Device End-To-End Model Surpassing Server-Side Conventional Model Quality and Latency
  11. Phoneme-Based Contextualization for Cross-Lingual Speech Recognition in End-to-End Models
  12. Better Morphology Prediction for Better Speech Systems
  13. Model Unit Exploration for Sequence-to-Sequence Speech Recognition
  14. Phoebe: Pronunciation-aware Contextualization for End-to-end Speech Recognition
  15. Lingvo: a Modular and Scalable Framework for Sequence-to-Sequence Modeling
  16. Dictionary Augmented Sequence-to-Sequence Neural Network for Grapheme to Phoneme prediction
  17. Sequence-to-Sequence Neural Network Model with 2D Attention for Learning Japanese Pitch Accents
  18. Pronunciation learning with RNN-transducers
  19. Learning Personalized Pronunciations for Contact Name Recognition
  20. NN-grams: Unifying neural network and n-gram language models for speech recognition
  21. On the compression of recurrent neural networks with an application to LVCSR acoustic modeling for embedded speech recognition
  22. Optimizing OPC data sampling based on orthogonal vector space
  23. Exploring the Nature of Trader Intuition
  24. Model-based scanner tuning in a manufacturing environment
  25. Investigating signal integration with canonical correlation analysis of fMRI brain activation data
  26. Human imagination in financial markets with insiders
  27. A Mind for the Market: an fMRI Study of Attribution of Mental States to Financial Markets
  28. SCR Recording During fMRI Acquisition


  1. Phoneme-based contextualization for cross-lingual speech recognition in end-to-end models
  2. Contextual biasing for speech recognition using grapheme and phoneme data
  3. Compressed recurrent neural network models
  4. Date and/or time resolution
  5. Learning personalized entity pronunciations
  6. Information matrix creation and calibration test pattern selection based on computational lithography model parameter
  7. Calibration pattern selection based on noise sensitivity
  8. Harmonic resist model for use in a lithographic apparatus and a device manufacturing method

Note: Additional US patent applications under review (closed to the general public).