Keynote Speakers for ICEMT 2020

Prof. Douglas O’Shaughnessy

Institut national de la recherche scientifique (INRS-EMT), Quebec, Canada

Adjunct Professor at McGill University, Canada

Fellow of IEEE

DOUGLAS O'SHAUGHNESSY received the B.Sc. and M.Sc. degrees in 1972 and the Ph.D. degree in 1976 from Massachusetts Institute of Technology, Cambridge, MA. I have been a professor at INRS-EMT (University of Quebec) in Montreal, Canada, since 1977. For this same period, I also taught every year as Adjunct Professor at McGill University in the Department of Electrical Engineering. I have worked as a teacher and researcher in the speech communication field for more than 40 years in the areas of automatic speech synthesis, analysis, coding and recognition.
I am a Fellow of the IEEE (2006), of the Acoustical Society of America (1992), and of ISCA (the International Speech Communication Association) (2019). I was the Chair of the IEEE SPS Speech and Language Processing Technical Committee (2013-2014; with member terms also: 1984-85, 2007-09 and 2011-12). I am a member of the Editorial Board of the IEEE Signal Processing Magazine (associate editor, 2012-2017), for EURASIP Journal on Advances in Signal Processing (2008-2020) and for JASA Express Letters (2009-2020). I was Secretary and an elected member of the ISCA Board (2009-2017). I was the founding Editor-in-Chief, EURASIP Journal on Audio, Speech, and Music Processing (2005-2015).
I am currently a member of the IEEE Fellow Committee (2018-2020), as well as its Fellow Strategic Planning Subcommittee (2020). I am a Member-at-Large, SPS Board of Governors (2019-2021), and on its Long-Range Planning Subcommittee (2020).
Prior service: Member, IEEE TAB Periodicals Committee (2011-2013); Member, IEEE Tellers Committee (2018-2019); elected a Regional Director of SPS, 2014-2015; Member, Technical Committee on Speech of the Acoustical Society of America (1995-97); Associate Editor, IEEE Transactions on Speech and Audio Processing (1995-99), Journal of the Acoustical Society of America (1998-2010); Member, SPS Conference Board (2000-05); Member-at-Large, SPS Board of Governors (2002-04; non-voting member, 2014-2015); and General Chair, 2004 International Conference on Acoustics, Speech, and Signal Processing.
I am the author of the textbook, Speech Communications: Human and Machine (IEEE Press, 2000). I was co-author, with Li Deng, of the book Speech Processing: A Dynamic and Optimization-Oriented Approach (Marcel Dekker, 2003). I presented tutorials on speech recognition at ICASSP ’96, ICASSP ’01 and ’09, and ICC ’03.

Speech Title: Automatic speech recognition (ASR): History, Progress, and Future

Abstract: In multimedia applications, speech is often one of the major means of communication. Since the advent of artificial intelligence (AI), machine recognition of speech has been viewed as a major challenge and application area. In recent years, ASR has been regularly found in many interactions between people and machines. However, its performance is still far from the level of how humans perceive speech, especially in adverse conditions. This talk will review how ASR works, how it developed, where its weaknesses still lie, and what we can expect in the future.
ASR is a pattern recognition task, but quite different from that of image recognition (which has led the recent explosion of machine learning technology). Images have inherent characteristics (edges, shapes, color) directly related to their nature (their “information”). Speech instead is highly encoded: ideas -> neural commands -> vocal tract shape and motion -> airflow -> resonances -> air-pressure modulation, each with its non-linear mapping. We will briefly review relevant aspects of human speech production and perception, then examine ways to analyze speech into sets of parameters and acoustic features, such as formants, energy, Fourier transform, Linear Prediction Coding, mel-frequency cepstrum, and periodicity.
Ways to convert these parameters to text are several: expert systems, dynamic time warping, Hidden Markov Models, and Deep Neural Networks are the main ones. We will examine the pros and cons of each of these, with regard to: accuracy, cost, real-time operation, and robustness to adverse conditions. All of these create models of speech representations at different levels: phonemes, syllables, words, sentences. Ease of training and adaptation of the models is a major factor in their usage. Ability to combine various sources of information for ASR is key, as human listeners decipher speech by bringing to bear far more than just the simple acoustic analysis of the inner ear, such as using learned models of the language being spoken.
Neural networks have shown strong ability to recognize major details of both static images and dynamic video of moving objects, but these obey fundamental principles of mass and motion, which do not apply easily to dynamic speech.

Prof. Akihiko Sugiyama

Yahoo! JAPAN Research

Fellow of IEEE and IEICE

Akihiko Sugiyama (a.k.a. Ken Sugiyama), affiliated with Yahoo! JAPAN Research, has been engaged in a wide variety of research projects in signal processing such as audio coding and interference/noise control. Prior to Yahoo Japan, he had a long career at NEC Corporation as a research engineer. His team at NEC developed the world's first Silicon Audio in 1994, a precursor of iPod. He served as the Chair of Audio and Acoustic Signal Processing Technical Committee, IEEE Signal Processing Society (SPS) [2011-2012], as associate editors for several journals such as IEEE Trans. Signal Processing [1994-1996], as the Secretary and a Member at Large to the Conference Board of SPS [2010-2011], as a member of the Awards Board of SPS [2015-2017], as the Chair of Japan Chapter of SPS [2010-2011], and a member of IEEE Fellow Committee. He was a Technical Program Chair for ICASSP2012. He has contributed to 17 chapters of books and is the inventor of 217 registered patents with more pending applications in the field of signal processing in Japan and overseas. He received 19 awards such as the 2002 IEICE Best Paper Award, the 2006 and 2018 IEICE Achievement Award, and the 2013 Ichimura Industry Award. He has delivered 145 invited talks in 75 cities of 27 countries. He is Fellow of IEEE and IEICE, and a Distinguished Lecturer for IEEE SPS [2014-2015] and for IEEE CE (Consumer Electronics Society) [2017-2018], and a Distinguished Industry Speaker for IEEE SPS [2020-2021].

Speech Title: History of Personal Media Terminals: From Walkman to Apple Watch

Abstract: A brief history of personal media terminals, highlighting the development of the Silicon Audio, the world’s first all solid-state audio player. The background of its development, its concept, and details of early versions are explained. The family of personal media terminals are presented followed by the impact on the following products such as smartphones, tablet PCs, and smart watches.

Prof. Kuan-Chou Chen
Purdue University Northwest, USA

Kuan-Chou Chen is the Associate Dean for Graduate Program and Research, Thomas M. McDermott Sr. Endowed Chair, Professor in Economic Development, Professor of Management Information Systems. He was the Department Head of Information Systems, Finance, and Business Analytics (2005-2016), as well as Interim Department Head of Department of Graduate Studies in Education (2013-2014) at Purdue University Northwest. He received his Ph.D. from Michigan State University and his MBA from National Cheng-Kung University in Taiwan. He specialized in computer programming, system simulation, project management, decision support systems, data mining, system analysis and design, e-business strategy and application, supply chain management, network design and security, knowledge management, and information economy. Professor Chen has more than 90 scholarly publications, most in peer-reviewed journals. He is an active participant in several professional journals and serves on three paper reviewer boards. Currently he is an Editor-in-Chief of International Journal of e-Education, e-Business, e-Management and e-Learning. His productivity and scholarship have been recognized by his colleagues, being nominated three years in a row for an “Outstanding Scholar Award.” He also the recipient of Teacher of the Year Award (Purdue University Northwest, 2005).

Speech Title: Current Challenges and Issues of Teaching and Learning in Higher Education

Absrtact: Today all kinds of challenge and issues teaching and learning in higher education are big topics of discussion, both in the news media and among the general public. The current higher education system is beset by a wide range of challenges, from declines in enrolments, demographics, technology trends to changes in disciplinary areas—and much more. In the meantime, for the future of universities and colleges around the world, there are also external, global factors that can change the trajectory of the industry. The educators in higher education system are all too aware it's not just domestic trends that can impact the way they operate their institutions.
Traditionally, colleges and universities have relied on standard economic models to sustain them. For most private institutions, that meant enrolling a stable number of tuition-paying students. In the case of public institutions, it meant receiving consistent state appropriations, in addition to tuition revenue. The pace of change in the economy, as well as the global COVID-19 pandemic, has impacted the reliability of traditional models, putting pressure on institutions to readjust their strategies.
Everyone agrees that providing high-quality education for our students is a worthy ideal.
However, there are many diverse viewpoints about how that should be accomplished. To that extend, understanding teaching and learning in higher education challenges and issues are important for teacher and students, and administrators. By being well-informed, they can contribute valuable input to the discussion. They can also make better decisions about what causes they will support or what plans they will make for their future.
This presentation provides detailed information on today's most relevant teaching and learning challenges and issues in higher education. It also outlines four emerging trends that have the potential to shake up the education sector.

Prof. Yiu-Wing Leung

Hong Kong Baptist University, Hong Kong

Yiu-Wing Leung received his B.Sc. and Ph.D. degrees from the Chinese University of Hong Kong, Hong Kong. Now he is Professor of Computer Science in the Hong Kong Baptist University, Hong Kong. On the teaching side, he has rich university teaching experience. He has received a number of teaching awards, including the President’s Award for Outstanding Performance in Teaching in 2011. He also has rich experience in curriculum design and development. He has been the Programme Director of an MSc in IT Management programme for fifteen years, leading its development and expansion. This programme is now the largest postgraduate programme in his university. At the same time, he has been serving as external examiners for local and overseas universities as well as accreditation organizations. On the research side, his research interests include computer communication, Internet computing and cloud computing. He has published more than 100 research papers and some of these papers are highly cited.

Speech Title: Teaching Pedagogy for Handling Diversity in Large University Classes

Abstract: Large university classes typically involve two diversity problems: (1) some students have good learning motivation but some students may not, and (2) some students have good learning capability but some students may not. It is important to properly tackle these diversity problems in order to achieve good learning outcomes. In this presentation, I will share some teaching methods for tackling these diversity problems. To tackle the diversity in learning motivation, daily life analogies could be used to illustrate the “dull” academic contents. A daily life analogy is a daily life example which has the same principle as the academic principle to be taught and it can easily be understood based on daily life experience. Using suitable daily life analogies, it would be easier to attract students’ attention, arouse their learning interests, and give a deep impression to them. To handle the diversity in learning capability, it would be very helpful to progressively explain from simple to complex, from concrete to abstract, and from high-level ideas to low-level details. In this manner, all students could progressively realize their reachable learning outcomes. I will present examples to illustrate how these teaching methods could be applied in practice to handle the diversity problems.

Prof. Mido Chang

Florida International University, USA

Dr. Mido Chang's research deals with statistical issues of Hierarchical Linear Models, Survival Analysis, Structural Equation Models, and Longitudinal Growth Models. She also applies the statistical models to a wide range of issues associated with the provision of equitable access to educational resources for all students, including linguistic and racial minority students. Her research findings support efforts to improve educational practice and policies for school programs, teacher instructional practices, and parent involvement. Dr. Chang's research projects have been funded from the Discovery Research K-12 Program of the National Science Foundation (NSF) and from the U.S. Department of Health and Human Services. She has served on review panels for national and international funding agencies and professional journals.

Speech Title: Collaborative Learning in Remote Classrooms
Abstract: COVID-19 brought dramatic changes and disruptions to our life, including education in which the remote instruction with technology becomes a-must. For unprepared teachers and instructors for this emergency education, researchers have offered practical, essential suggestions. Among them, the salient tips can be summarized as, try to understand students’ needs, organize the course material intuitively, add visual materials, explain the learning expectation, scaffold learning activities, provide examples, make the remote class comfortable, commit to continuous improvement, etc.
Along with those new challenges to improve students’ remote learning, researchers highlight the importance of a socio-constructive aspect of education: inviting students to cyber collaboration among themselves rather than only focusing on instructional activities of instructors. Many researchers have well evidenced the benefits of collaborative learning in face-to-face classrooms. Through collaborative learning, students gain a greater understanding of a topic, more in-depth knowledge, higher-level engagement to a subject, and increased motivation to learn. Collaboration indicates the mutual commitment of students in a coordinated effort to solve the problem together rather than dividing work among them. Therefore, in the environment of collaborative learning, students work together in a small group to solve a problem, helping each other. The best part of the collaborative learning environment is to nurture students to uncover the unifying core concept and organize topics with the guiding principle in a healthy educational setting.
How can we achieve the best outcomes with collaborative learning in our remote classrooms? Research provides several suggestions: 1) Make sure computer-mediated networks support for social interaction and collaboration, 2) Focus on what is uniquely feasible with new technology, 3) Apply what we know about collaborative learning in the face-to-face classrooms, 4) With regards to the group size, groups of three work effectively to enable each student to participate fully and to build group cohesion, 5) For the group composition, heterogeneous groups in terms of gender, status, culture, or expertise are more productive in integrating diverse ideas, 6) structuring of group activities is required to avoid information overload. Still, too much scripting leads to less interaction, 7) Let students create goals that jointly agreed upon, 8) Help students to build meaning-making that brings out during social negotiation of a learning task or object, etc.
I will open further discussion about unprecedented suggestions to achieve deeper learning through collaborative learning in remote classrooms.