Qiujia Li (李秋佳)

Second-year PhD Student
Machine Intelligence Laboratory
University of Cambridge

Email: ql264 [at] cam (dot) ac (dot) uk

Research interests:

  • automatic speech recognition
  • speech processing
  • machine learning

I am funded by Peterhouse Graduate Studentship.

Education

University of Cambridge

Master of Engineering, Bachelor of Arts (Hons.)
Distinction, Information and Computer Engineering
Oct. 2014 — June 2018, Cambridge, UK

Massachusetts Institute of Technology

Cambridge-MIT Exchange
GPA: 5.0/5.0, Electrical Engineering and Computer Science
Sept. 2016 — May 2017, Cambridge, MA, US

Publications

Thesis

Confidence Scores for Speech Processing
Qiujia Li
Univerisity of Cambridge, 2018MEng Thesis

Preprints

Discriminative Neural Clustering for Speaker Diarisation
Qiujia Li*, Florian L. Kreyssig*, Chao Zhang, and Philip C. Woodland (* equal contribution)
arXivPaperCode

Conference Papers

Integrating Source-channel and Attention-based Sequence-to-sequence Models for Speech Recognition
Qiujia Li, Chao Zhang, and Philip C. Woodland
ASRU 2019Paper

Bi-directional Lattice Recurrent Neural Networks for Confidence Estimation
Qiujia Li*, Preben M. Ness*, Anton Ragni, and Mark J. F. Gales (* equal contribution)
ICASSP 2019PaperCodePoster

PyHTK: Python Library and ASR Pipelines for HTK
Chao Zhang, Florian L. Kreyssig, Qiujia Li, and Philip C. Woodland
ICASSP 2019PaperPoster

Confidence Estimation and Deletion Prediction Using Bidirectional Recurrent Neural Networks
Anton Ragni, Qiujia Li, Mark J. F. Gales, and Yu Wang
SLT 2018PaperPoster

Inverting Audio-visual Simulation for Shape and Material Perception
Zhoutong Zhang, Jiajun Wu, Qiujia Li, Zhengjia Huang, Joshua B. Tenenbaum, and William T. Freeman
CVPR 2017 WorkshopPaperProject Page

Shape and Material from Sound
Zhoutong Zhang, Qiujia Li, Zhengjia Huang, Jiajun Wu, Joshua B. Tenenbaum, and William T. Freeman
NIPS 2017 (spotlight)PaperProject Page

Generative Modeling of Audible Shapes for Object Perception
Zhoutong Zhang, Jiajun Wu, Qiujia Li, Zhengjia Huang, James Traer, Josh H. McDermott, Joshua B. Tenenbaum, and William T. Freeman
ICCV 2017PaperProject Page

Experimental Studies on Teacher-student Training of Deep Neural Network Acoustic Models
Qiujia Li, Chao Zhang, Florian L. Kreyssig, and Philip C. Woodland
UK Speech 2017Poster

Research Experience

Speech and Dialog Research Group @ Microsoft

Topic: speaker diarisation
Mentor: Jasha Droppo, Fil Alleva
July 2018 — Sept. 2018, Bellevue, WA, US

Machine Intelligence Laboratory (MIL) @ Univerisity of Cambridge

Topic: confidence scores for speech processing
Supervisor: Prof. Mark Gales
Oct. 2017 — June 2018, Cambridge, UK

Topic: teacher-student training of acoustic models
Supervisor: Prof. Phil Woodland
Summer 2016 & Summer 2017, Cambridge, UK

Computer Science and Artificial Intelligence Laboratory (CSAIL) @ MIT

Topic: visual and auditory perception
Supervisor: Prof. Bill Freeman, Prof. Josh Tenenbaum
Dec. 2016 — May 2017, Cambridge, MA, US

Miscellany

Sports: badminton, swimming, snooker/pool, basketball, and table tennis.

Hobbies: traveling, photography, and aviation.

Contact

Room BE5-06, Department of Engineering
Trumpington Street, Cambridge, CB2 1PZ, UK
Email: ql264 [at] cam (dot) ac (dot) uk