Center for Robust Speech Systems

John H.L. Hansen (IEEE: S'81-M'82-SM'93) was born in Plainfield, New Jersey. He received the B.S.E.E. degree with highest honors from Rutgers University, New Brunswick, N.J. in 1982. He received the M.S. and Ph.D. degrees in Electrical Engineering from the Georgia Institute of Technology, Atlanta, Georgia, in 1983 and 1988, respectively.

He is presently Department Chair and Professor in the Department of Electrical Engineering at the Erik Jonsson School of Engineering and Computer Science, University of Texas at Dallas. He holds the Endowed Chair in Telecommunications Engineering, and has a secondary appointment in the School of Brain and Behaviorial Sciences (Callier Center - Speech & Hearing). From 1999 until 2005, he was a Professor in the Departments of Speech, Language & Hearing Sciences (SLHS), and Electrical & Computer Engineering (ECE), and served as Department Chairman in SLHS at the University of Colorado Boulder. In 1988, he established the Robust Speech Processing Laboratory (RSPL), and directed the Robust Speech Processing Group (RSPG-CSLR). He was co-founder of the Center for Speech and Language Research (CSLR) where he served as Associate Director. He was a faculty member at Duke Univ., Departments of Electrical and Biomedical Engineering for eleven years before joining Univ. of Colorado in 1999. Previously, he was employed by the RCA Solid State Division, Somerville, N.J., (1981-82), and Dranetz Engineering Laboratories, Edison, N.J., (1978-81). He has served as a technical consultant to industry and the U.S. Government, including AT&T Bell Laboratories, I.B.M., Infoture, BB&N, Sparta, ASEC, VeriVoice, HRL, and DOD in the areas of voice communications, wireless telephony, robust speech recognition, and forensic speech/speaker analysis. His research interests span the areas of digital speech processing, analysis and modeling of speech and speaker traits, speech pathology and voice assessment, speech enhancement and feature estimation in noise, robust speech recognition with current emphasis on robust recognition and training methods for topic spotting in accent, noise, stress, and Lombard effect, and speech feature enhancement in hands-free environments for human-computer interaction.

EDUCATION

Ph.D. Electrical Engineering

• Georgia Institute of Technology, July, 1988.
• Areas of concentration: Digital Signal Processing, Speech Processing, Communications.
• Thesis title: Analysis and Compensation of Stressed and Noisy Speech with Application to Robust Automatic Recognition.
• Thesis advisor: Mark A. Clements.
• Thesis committee chairman: Ronald W. Schafer

M.S. Electrical Engineering

• Georgia Institute of Technology, (with highest honors), December 1983.
• Areas of concentration: Digital Signal Processing, Voice Communications.

B.S. Electrical Engineering

• Rutgers University, College of Engineering, (with highest honors), May, 1982.
• Areas of concentration: Digital Signal Processing, Communications, and Biomedical Engineering.

EXPERIENCE

University of Texas at Dallas

• Department Chair: Department of Electrical Engineering, 8/05 - present.
• University Distinguished Chair in Telecommunications Engineering, 8/05 - present.
• Director and Founder: Center for Robust Speech Systems(CRSS), 8/05 - present.
• Professor: Department of Electrical Engineering, Erik Jonsson School of Engineering and Computer Science, 8/05 - present.
• Professor: Speech and Hearing (Callier Center), School of Behavorial and Brain Sciences, 8/05 - present.

University of Colorado at Boulder

• Department Chair: Department of Speech, Language, and Hearing Sciences, 8/03 - 8/05
• Director: Robust Speech Processing Group(RSPG), 1/99 - 8/05
• Co-Founder: Center for Spoken Language Research(CSLR), Associate Director (1/99 - 9/03)
• Professor: Department of Speech, Language, and Hearing Sciences, 5/04 - present.
• Professor: Department of Electrical and Computer Engineering, 5/04 - present.
• Associate Professor: [1/99 - 5/04] (with tenure)

Duke University

• Director, Robust Speech Processing Laboratory(RSPL), 8/88 - 12/98.
• Associate Professor. [5/96 - 12/98.]
• Assistant Professor. [8/88 - 5/96.]
• Department of Electrical and Computer Engineering, 8/88 - 12/98.
• Department of Biomedical Engineering, 1/94 - 12/98.

RESEARCH:

Founded The Center for Robust Speech Systems at the Univ. of Texas at Dallas in 2005 which is focoused on interdisciplinary research in speech processing and human computer interaction. Originally founded the Robust Speech Processing Laboratory in 1988 at Duke University, which became the Robust Speech Processing Group at CSLR in 1999, Univ. of Colorado, now the Center for Robust Speech Systems (CRSS-UTD). Research has focused on interdisciplinary areas in robust speech processing and recognition, digital signal processing, speech and speaker analysis, human-computer interfaces/interaction, speech enhancement and processing in noise, and voice communications. Specific research accomplishments include analysis and modeling of speech for voice and telecommunications applications, robustness issues for speech recognition in adverse conditions, speech enhancement, spoken document retrieval, in-vehicle voice navigation dialog systems, and voice modeling in noise, stress, accent, and Lombard effect. Has also had extensive collaborations on research in telemedicine, resulting in numerous contributions in medical speech processing involving detection and quality assessment of speech under stress and pathology. At Univ. of Colorado, he co-founded the Center for Spoken Language Research, which is now CRSS at UTD and has focused on rich transcription for spoken document retrieval, hands-free robust recognition for in-vehicle voice navigation, advanced human-computer interaction, speaker/speech analysis for stress, accent/dialect analysis and modeling, and Lombard effect. While at Duke University Dept. Electrical Engineering, established collaborative research program with Duke Medical Center (Dept. of Pediatrics, Speech & Language Pathology), Johns Hopkins Medical Center, Scottish Rite Children's Medical Center. At University of Colorado, he has established collaborations with Walter Reed Army Institute of Research, NATO ITO/RSG.10, Nagoya University (Japan), Aalborg University (Denmark), Helsinki Univ. of Technology (Finland), Vincent Voice Library (Michigan State Univ.), Collaborative Digitization Program (Colorado), and Tufts University Digital Library.

The majority of research in the field of speech processing for telecommunications focuses on formulating algorithms for ideal laboratory conditions. As many have shown, the resulting algorithms do not perform well in real or actual voice communication environments. The systems we develop focus on addressing issues relating to robustness issues of speech applications in telecommunications. Some of these issues relate to speech produced in noise, high stress, accent, or Lombard effect. The future of Multimedia human/machine interfaces will depend on system approaches which are capable of addressing the wide range of adverse conditions in which speech may be produced for applications in speech recognition, communications, language identification, speaker recognition, and other telecommunication applications.

Formulated the first class of constrained iterative speech enhancement algorithms based on novel properties of speech production and auditory processing. Established the only working framework based on Source Generator Theory for characterizing speech production under stress, resulting in novel stress equalization and noise suppression algorithms for robust speech recognition in noise, stress, and Lombard effect. NSF funded collaborative research program (with Michigan State Univ.: Electrical Engineering [Prof. Deller] and Vincent Voice Library [Prof. Seadle]) resulted in the development of SpeechFind [http://Speechfind.utdallas.edu]

--- spoken document retrieval system which spans audio from the past 110 years. Research also in the area of smart vehicle systems for improved human-vehicle interaction. Includes DARPA sponsored CU-Move and NEDO/UTD supported UTDrive [http://www.utd.edu/research/utdrive/Home.htm], and international collaboration with Japan, Turkey, Singapore, Italy.

The Robust Speech Processing Laboratory was established in 1988, and continued on as the Robust Speech Processing Group at CSLR, and now CRSS at UD-Dallas.. Has served as Principal Investigator on more than $12M in external funded research from such sources as NSF, Whitaker Foundation, DoD/DARPA, U.S. Air Force, U.S. Army, U.S. Navy, TSWG, IBM (T.J. Watson Research Center, Voice & Telephony; MWave DSP Development), AT&T Bell Labs (Government Systems, Wireless), Motorola, HRL, SpeechWorks, Visteon/Ford, Voice Signal Technologies, Mishubishi, Toyota CR&D, and others. Author of more than 250 journal and conference papers and 7 textbooks in the area of speech and signal processing, and is coauthor of the textbook Discrete-Time Processing of Speech Signals, IEEE Press, 2000, DSP for In-Vehicle and Mobile Systems, Springer, 2004, and the research monograph Automatic Recognition of Speech in Stressful Environments, published by DoD, 1991, and was lead author of the monograph "The Impact of Speech Under "Stress" on Military Speech Technology," NATO Research and Technology Organization, RTO-TR-10 AC/323(IST)TP/5 (ISBN 92-837-1027-4).

Contact

E-mail: John H.L. Hansen
Office: 972-883-2910
Fax: 972-883-2710
http://crss.utdallas.edu/jhlh/