Prof. MENG, Mei Ling Helen 蒙 美 玲 教授
Professor and Chairman
SB (MIT, USA)
|Office:||Room 609B, William M.W. Mong Engineering Building|
|Phone:||(852) 3943-8316, 3943-8327|
Helen M. Meng is Professor in the Department of Systems Engineering and Engineering Management. She was also the immediate past Associate Dean of Research in the CUHK Faculty of Engineering. She received her Bachelor of Science, Master of Science and Doctor of Philosophy degrees in Electrical Engineering, all from the Massachusetts Insitute of Technology. She then worked as a Research Scientist in the MIT Laboratory for Computer Science Spoken Language Systems Group. She joined CUHK in 1998 and established two research laboratories -- the Human-Computer Communications Laboratory in 1999 and the Microsoft-CUHK Joint Laboratory for Human-centric Computing and Interface Technlogies in 2005. She current serves as Director and Co-Director for these two laboratories respectively.
Helen is Principal Investigator in a series of projects funded by government and industry, including two multi-million dollar projects on mobile, multi-modal and multilingual (M3) computing and on multi-device access to Web content. A brief listing of her projects is provided below.
Helen's interests in speech and language research cover areas with basic research challenges as well as exciting practical potentials. Her approaches strive for theoretic grounding on acoustic-phonetics, phonology, linguistics, signal processing, pattern recognition, artificial intelligence and machine learning. Her three main research perspectives are:
- Spoken language as a Human-centric Interface: Enabling technologies include speech recognition and synthesis, language understanding/generation and dialog modeling, which can be integrated into a spoken dialog system. Helen's recent emphasis is on audio-visual, multimodal processing, e.g. expressive talking avatars, semantic integration of input speech and gestures for ubiquitous computing, etc.
- Spoken language as Conversational Biometrics: This characterizes speakers in terms of their vocal tract characteristics, speaking styles and personalized linguistic content. Enabling technologies include speaker authentication, combinable with utterance verification, lip-tracking, facial image and fingerprint verification for multimodal biometric authentication. Helen's recent emphasis is on decision fusion strategies in multi-biometric authentication and its non-intrusive integration with a conversational dialog interface.
- Spoken language as Media Content: Mining digital media needs efficient, automatic techniques for indexing, searching and organizing large stores of text, audio and video soundtracks. Enabling technologies include speech detection, speech recognition, information retrieval, extraction and summarization. Helen's recent emphasis is on prosodic and lexical analyses of broadcast speech.
Helen is Elected Member of the IEEE Speech Technical Committee and Conference Board Representative; Editorial Board Member in Speech Communication, Computer Speech & Language and the International Journal on Computational Linguistics and Chinese Language Processing; Member of the W3C VBWG promoting internationalization of the Speech Synthesis Markup Language; and External Examiner of Ph.D. theses in Tsinghua University, National University of Defense Technology and Universite Paris XI. She also serves as appointed member of the HKSAR Government Digital 21 Strategy Advisory Committee and Working Group on Chinese Information Processing. She has given invited talks in many conferences and serves on their scientific review committees. She is former Chairperson of the ACM Hong Kong Chapter and an Executive Committee Member of the IEEE Signal Processing Society (HK Chapter). She is also Guest Professor of the Chair Professor Project in Tsinghua University's Department of Computer Science and Technology, as well as in Fudan University's Key Laboratory in Intelligent Information Processing.
In terms of education, student supervision and services, Helen established CUHK's worldwide Engineering Student Exchange Program and received the Faculty of Engineering Service Award 2005. Her Ph.D. advisee was elected Microsoft Fellow 2005. In previous years, she received the Faculty of Engineering Exemplary Teaching Award and her advisees have won the Microsoft Imagine Cup Hong Kong Championship, three Challenge Cup Prizes and the Faculty of Engineering旧 Outstanding Thesis Award. She supervised 26 Masters and Ph.D. students has over 100 publications.
In addition to human speech and language research, Helen is also interested in the development of Information and Communication Technologies in Hong Kong, the Asia-Pacific and other areas. She serves as appointed member of the HK SAR Government Task Force on Facilitating the Adoption of Wireless and Mobile Services and Technology and the Central Committee on IT for Rehabilitation Services. She is also Council Member of the Hong Kong Computer Society. She co-authored the Information Technology sector report of the book, "Made By Hong Kong", which topped the South China Morning Post best-seller's list in 1997.
Recent/Upcoming Professional Activities
- Tutorial and Plenary Chair, International Symposium on Chinese Spoken Language Processing Singapore, December 2006.
- Area Chair, IEEE International Conference on Acoustic, Speech and Signal Processing, Toulouse 2006
- Far East Liaison, Interspeech 2005, Lisbon, September 2005
- Area Chair, Human Language Technlogy Conference, Vancouver, December 2005
- Vice-Chair, International Symposium on Chinese Spoken Language Processing Hong Kong, December 2004.
- Invited Talk, "Spoken Language Technologies and Machine Translation, 20th Anniversary Conference of the CUHK Department of Translation, Hong Kong, December 4, 2004.
- Program Committee Member, Hong Kong International Computer Conference 2004, November 2004.
- Invited Talk, "The Author Once, Present Anywhere (AOPA) Software Platform: Towards Universal Accessibility for Chinese Web Content Development," Customer Contact World, Kuala Lumpur, Malaysia, April 2004.
- General Co-Chair, International Speech Communication Association, Workshop on Multilingual Spoken Document Retrieval, Hong Kong and Macau, April 4-6, 2003.
- Invited Talk, "Spoken Language Technologies in Multilingual, Multimodal and Multimedia Systems," Voice World 2003, 12-13 March, 2003, Suntec Singapore.
- Invited Talk, (Title to be determined) City University of Hong Kong, Department of Electronic Engineering seminar, March 6, 2003.
- Invited Talk, "The AoE-IT Web Repository: Engendering Technologies in Multimedia Retrieval and Multilingual, Multi-document Summarization," Hong Kong University of Science and Technology, Department of Computer Science seminar, February 10, 2003.
- Invited Talk, "Intelligent Speech for Information Systems (ISIS): A Multimodal, Trilingual, Distributed Conversational System with Combined Interaction and Delegation Dialogs," HANPA, Tsinghua Unviersity,, Beijing, December 8, 2002.
- Invited Talk, "Intelligent Speech for Information Systems (ISIS): A Multimodal, Trilingual, Distributed Conversational System with Combined Interaction and Delegation Dialogs," International Symposium on Chinese Spoken Language Processing, Taipei, August 23, 2002.
- Invited Panelist, International Speech Communication Association Workshop on Pronunciation Modeling and Lexicon Adaption for Spoken Language, Estes Park, Colorado, September 13, 2002.
- Invited Talk, "Research and Development on Speech and Language Technologies at the Chinese University of Hong Kong," Voice Recognition Seminar, Hong Kong SAR Government Information Technology Services Department, May 10, 2002.
- Keynote Speech, "The Use of Belief Networks for Mixed-Initiative Dialog Modeling," Research on Computational Linguistics Speech Signal Processing Conference, Academia Sinica, Taipei, April 26, 2002.
- Invited Talk, Special Session on Speech Synthesis in Minan, Cantonese and Hakka Dialects, Research on Computational Linguistics Speech Signal Processing Conference, Academia Sinica, Taipei, April 26, 2002.
Helen is the recipient of the Department's Exemplary Teaching Award 2001.
- SEG 5131 Multimedia and Multimodal Processing
- SEG 5130 Spoken Language Systems
- SEG 3510 Human-Computer Interaction
- SEG 2430 Applied Probability and Statistics
- SEG 5010 Advanced Database and Information Systems (course coordinator: Professor K. F. Wong)
- ECT 7250/7150 Projects in E-Commerce Technologies
- Multimodal Interfaces and Multi-biometric Authentication
- Multilingual Media Mining / Multimedia Retrieval
- Multilingual Spoken Dialog Systems
- Natural Language Understanding
- Speech Recognition and Generation
- Intelligent agents & computer-supported collaborative work
- Human-Computer Interaction
HCCL's Research on Multimodal Processing
Helen is PI of a project that aims to develop mobile, multimodal and multilingual (M3) computing technologies. We are collection and annotating the M3 corpus that consists of speech, facial images and fingerprints collected from a large number of subjects with destop computers, PDAs and 3G phones. We are developing multi-biometric authentication technologies, combined with utterance verfication technology.
Another project in multimodal processing is the development of an expressive avatar, named LinLin. Please check out this link.
CU VOCAL: A Corpus-based Approach for Highly Natural Cantonese Text-to-Speech Synthesis
Helen is the PI of the CU VOCAL project. Our original objective is to develop a domain-optimized speech generation engine across Chinese dialects (focusing on Cantonese and Putonghua) as a component technology in Spoken Dialog Systems. We have successfully designed and implemented a novel approach for this task, and applied it to application domains such as foreign exchange, stocks and finanical news. The latest development of the CU Vocal project is that we are now able to synthesize speech for domain-independent text for Cantonese. The naturalness of the synthesized speech is our main concern. Feel free to test out our latest demo (released August 2001) and give us feedback! See CU VOCAL homepage
Project MEI: Mandarin-English Information -- Investigating Translingual Speech Retrieval
Helen is team leader of Project MEI at the Johns Hopkins University Summer Workshop 2000, sponsored by the National Science Foundation in the US. The team developed one of the first English-Chinese cross-media and cross-language spoken document retrieval systems. English text queries are used to retrieve Mandarin audio documents. MEI is a collaborative project across eight institutions. See MEI homepage
ISIS (Intelligent Speech for Information Systems): A Trilingual Speech Interface for Financial Information Access
We are developing a spoken dialog system for the stocks domain, which can handle the languages of Hong Kong -- English, Cantonese and Putonghua. ISIS integrates a plethora of speech and language technologies, including trilingual recognition and synthesis, natural language understanding, dialog and discourse modeling, speaker verification, and the use of software agents for asynchronous human-computer interaction. This is a collaborative project with Peking University's National Key Laboratory for Machine Perception, under the Joint Center for Intelligence Engineering between CUHK and PKU. See ISIS homepage
CU FOREX: A Bilingual (English-Cantonese) Conversational Hotline for Real-Time Foriegn Exchange Inquiries
Helen is the principal investigator of the CU FOREX project, sponsored by SpeechWorks International Ltd., IVRS (International) Ltd., and Reuters Hong Kong. Please call and try out system! See CU FOREX homepage.
On-line Auctions with Intelligent Agents
This groupware supports on-line auctions which can simultaneously involve both real users and intelligent software agents. The system has proven effective for the allocation of projects among student teams in one of our courses.
Check out our demos!
Please click here for the listing. An Invitation... Come visit our Human-Computer Communications Laboratory!