Ruili Wang B.E.
M.E. PhD (Dublin)
|
|
Senior
Lecturer Computer
Science and Information Technology Private
Bag 102-904, North Shore Mail Centre Auckland,
New Zealand Phone:
(+64) 09-4140800 ext 43258 (from Auckland) Fax: +
64–6–350 2259 Mobile: ask me Free Phone: 0800-Massey (ext. 43258) |
|
Curriculum Vita | Publications | Projects | Scholarships
|Research | Click onto Science | PhD thesis | Photos |
A Full Marsden Fund $660,870 (plus GST)
“No Longer Lost in Recognition: Development of
Novel Large‐Vocabulary
Continuous Speech Recognition Techniques”
(2014-2016)
Speech Processing and
Maori Speech Recognition
Speech
recognition is one of the main areas in speech processing and signal
processing. Large-vocabulary continuous speech recognition (LVCSR), especially
spontaneous speech recognition, remains a grand scientific challenge in the signal
processing and artificial intelligence communities. This research aims to
develop several novel LVCSR techniques. We will then apply our newly developed
techniques and other state-of-the-art speech recognition technologies to
develop a Maori speech recognition system that can convert Maori speech (such
as continuous news broadcast speech and spontaneous speech) to Maori text. This
research will for the first time develop a large-vocabulary continuous Maori
speech recognition system.
The
overall aims of the proposed research include: (i) development of novel
Automatic Speech Recognition (ASR) methods for large-vocabulary continuous
speech recognition (LVCSR); (ii) development of large-vocabulary continuous
Māori speech recognition systems based on the new LVCSR methods developed
in this project and state-of-the-art speech recognition technologies. Four
research objectives underpin this project:
Objective
1 will involve developing novel acoustic and language modelling methods.
Objective
2 will involve developing robust speaker adaptation methods.
Objective
3 will involve developing techniques for spontaneous speech recognition.
Objective
4 will involve developing Māori speech recognition systems (MSR) based on
the novel methods developed in this project and other state-of-the-art
technologies.
Principle
Investigator: Dr.
Ruili Wang
Associate
Investigators:
·
Professor
Tatsuya Kawahara, Kyoto University JAPAN
·
Associate
Professor IR Lane Carnegie Mellon University USA
·
Associate
Professor A Lee, Nagoya Institute of
Technology JAPAN
·
Associate
Professor WH, Abdulla, The University of Auckland
·
Dr DJ Joseph
Massey University
Notes:
• Principal Investigator, Dr. Ruili
Wang, who will lead the research and supervise the overall project, has
expertise in speech processing and machine learning.
• Overseas Associate Investigators:
·
Prof.
Tatsuya Kawahara, Kyoto University, Japan
·
Associate
Prof. Ian Lane, Carnegie Mellon University, USA
·
Associate Prof.
Akinobu Lee, Nagoya Institute of Technology, Japan
All have
strong expertise in speech recognition, spoken language understanding, and
spoken dialogue systems.
• New Zealand Associate Investigator:
Associate Prof. Waleed Abdulla, Auckland University, who has strong expertise
in signal processing.
• New Zealand Associate Investigator: Dr
Darryn Joseph, Massey University, who is an expert in Te reo Māori.
News
·
$3m
in Marsden grants for Massey researchers
·
3 PhD
scholarships in Speech Recognition available