10 best universities in the field of speech technology and artificial intelligence

Over the past few years, interest in speech interfaces has reawakened in Russia. Western scientific tradition, unlike the Russian one, has in this direction continuous more than half a century of experience.
Our review is dedicated to leading universities that provide education in the field of speech technologies - automatic speech processing, voice interfaces, biophysics, artificial intelligence, neural networks, etc.

At the end of the list, a few words describe the situation with the Russian educational market in the field of speech technologies and two specialized departments at MIPT and ITMO.

Harvard University - Massachusetts Institute of Technology
Speech and Hearing Bioscience and Technology (SHBT)

“A strange but universally recognized fact - an idea, emotion, signal, song can travel from one person’s head to another’s head, and this movement depends on the frighteningly complex, unusually exciting chain reaction known as human communication. In the program of our university, we study every connection in this chain, at every level of knowledge, from biochemistry to understanding. ”

Since 1992, about 50 students from 60 different departments of Harvard University, the Massachusetts Institute of Technology, Boston University and Harvard Training Hospitals have been studying annually at SHBT.

Main research interests of SHBT:

• Basic research of the speech apparatus and speech functions.
• Clinical studies of the human voice and speech abnormalities.
• Mechanics, biophysics, physiology and / or molecular biology of the middle and inner ear.
• Acquired or congenital hearing impairment.
• Neurophysiological or modeling approaches in the study of nerve cells and the patterns underlying auditory processing.
• Neuro-visual studies of tinnitus mechanisms.
• Cognitive neurobiology of language signal processing.
• Design, development and improvement of the hardware and software system of hearing aids, ear implants, vestibular prostheses or automatic speech recognition algorithms.

Candidates for SHBT must have a bachelor's degree in physics, biology, psychology, linguistics, communication sciences, technology and computer science, and have extensive analytical skills.

reference

Institution : Harvard-MIT Division of Health Sciences and Technology
Direction : Program in Speech and Hearing Bioscience and Technology
Faculty : Speech and Hearing Bioscience and Technology
Website : web.mit.edu/shbt
Disciplines : Acoustical Signal Processing, Engineering Acoustics, Medical / Bioacoustics , Musical Acoustics, Physical Acoustics, Psychological Acoustics, Speech, Animal Bioacoustics
YouTube Channel : www.youtube.com/Harvard , www.youtube.com/MIT
Address : E25-518, 77 Massachusetts Avenue, Cambridge, MA 02139, USA [ on the map ]
Information for applicants :goo.gl/fOuAX
Contacts : 617-2537498 (fax), shbt-admissions@mit.edu

Stanford School of Engineering
Mechanical Engineering

“The future is limited only by our imagination, and the possibilities are endless.”

Stanford School of Engineering, founded in 1925 and located in the heart of Silicon Valley, annually accommodates 9 different departments, about 200 teachers and 4 thousand students. 65 laboratories, many of which are multidisciplinary, work in the fields of medicine, business, linguistics and physics.

Main research interests and areas of SOE:

• Aeronautics and astronautics.
• Bioengineering.
• Chemical technology.
• Civil and environmental engineering.
• Computer science.
• Electrical Engineering
• Management in the field of science and technology.
• Materials Science.
• Machine design.

reference

Institution : Stanford School of Engineering
Direction : Mechanical Engineering; ME & Aero. & Astro.
Website : soe.stanford.edu
Disciplines : Medical / Bioacoustics, Physiological Acoustics, Structural Acoustics and Vibration, Engineering Acoustics, Noise and Noise Control, Nonlinear / Aeroacoustics
YouTube Channel : www.youtube.com/StanfordUniversity
Address : Stanford, CA 94305, USA [ map ]
Information for applicants : goo.gl/PuYOY
Contacts : chasst@stanford.edu ( Charles R. Steele ), lele@stanford.edu ( Sanjiva K. Lele ), pinsky@stanford.edu (Peter Pinsky )

Cambridge University Engineering Department
The Machine Intelligence Laboratory

“The Speech Research Group is part of the Machine Intelligence Laboratory. The mission of SRG is to promote knowledge of spoken language machine processing and the development of efficient algorithms for implementing applications. The main specification of SRG is to work with large speech dictionaries and related technologies. Research interests also extend to conversational dialogue systems, pattern recognition, speech synthesis, and machine learning. ”

Main research interests and areas of SRG:

• Acoustic modeling (statistical models).
• Basic research in machine learning.
• Streamline dialogue using reinforcing learning.
• Recognition in large dictionaries.
• Pattern recognition.
• Speech recognition on mobile devices.
• Independence and noise reduction.
• Dialogue systems and VoiceXML.
• Statistical language modeling.
• Statistical machine translation.
• Processing and transcribing recognized speech.

Speech Research Group accepts applications from potential graduate students and doctoral candidates. 1 or 2 year graduate courses are also possible.

reference

Institution : Cambridge University, Speech Research Group
Direction : The Machine Intelligence Laboratory
Faculty : Cambridge University Engineering Department.
Website : mi.eng.cam.ac.uk/mi/Main/Speech
Disciplines : large vocabulary speech transcription, spoken dialogue systems, multimedia document retrieval, speech synthesis, machine learning.
YouTube Channel : www.youtube.com/CambridgeUniversity
Address : Trumpington Street, CB2 1PZ, UK [ map ]
Information for applicants : goo.gl/VbucH
Contact: 01223 332752 (tel.), 01223 332662 (fax), jrm16@eng.cam.ac.uk (Janet Milne)

University of Oxford
Speech & Brain Research Group

“We are interested in how the sensory and motor regions of the brain interact in speech communication. We use various methods of displaying brain activity to study the brain during acts of utterance and speech perception. "

Speech & Brain Research Group is recruiting potential masters and doctors who can choose any of the courses of the Department of Experimental Psychology.

Main research interests and areas of FMRIB:

• Analysis of functional and structural data of brain images.
• Physiological neuroimaging.
• Brain Disorders.
• Diffusion image.
• Speech and brain.
• Visualization.
• Neurodegeneration.
• Cognition.
• Psychiatry.
• Epilepsy.

reference

Institution : University of Oxford
Direction : Center for Functional Magnetic Resonance Imaging of the Brain (FMRIB); Speech & Brain Research Group
Faculty : Department of Experimental Psychology, Oxford Center for Developmental Science.
Website : www.fmrib.ox.ac.uk/speech-and-brain
Disciplines : brain structure, neural activity, emotional processing, non-speech stimuli.
YouTube Channel : www.youtube.com/Oxford
Address : Wellington Square, OX1 9FB Oxford, UK [ map ]
Information for applicants : goo.gl/HDMcO
Contacts: kate.watkins@psy.ox.ac.uk, +44 (0) 1865 280459 (tel.), +44 (0) 1865 280300 (fax)

University of California, Los Angeles (UCLA)
Department Of Linguistics

"UCLA Linguistics Department is one of the world's leading centers for the scientific study of language."

Main research interests and areas of UCLA LD:

• Phonetics.
• Phonology.
• Syntax.
• Semantics.
• Psycholinguistics.
• Matlingistics.
• Historical Linguistics.
• African, Native American languages.

There are laboratories for phonetics, psycholinguistics, and language training. List of linguistic disciplines .

reference

Institution : University of California, Los Angeles
Direction : Department Of Linguistics
Website : www.linguistics.ucla.edu
Disciplines : phonetics, phonology, syntax, semantics, psycholinguistics, language acquisition, historical linguistics, mathematical linguistics.
YouTube Channel : www.youtube.com/UCLA
Address : 3125 Campbell Hall, UCLA, Los Angeles, USA [ map ]
Information for applicants : goo.gl/cdvYF
Contacts : (310) 825-0634 (tel.), + (310) 206-5743 (fax), linguist@humnet.ucla.edu

Johns Hopkins University
The Center for Language and Speech Processing

“Automated systems that interact with people through conversation or writing will soon increase their usability, ease of use, and therefore our productivity. These systems will accompany us wherever information is found, and everyone, including people with disabilities, will be able to access large and unstructured databases, such as the Internet. ”

Center for Language and Speech Processing (CLSP) was established in 1992 with the support of the US government (NSF, DARPA, DoD). The studies are conducted by teachers, researchers and graduate students affiliated with six related faculties: bioengineering, cognitive science, computer science, electrical and computer engineering, mathematical sciences and psychology.

Key research interests and areas of CLSP:

• Language modeling.
• Natural language processing.
• Neural processing.
• Acoustic processing.
• Optimization theory.
• Language entry.

CLSP accepts undergraduate and graduate students. Applications must be submitted through any of these faculties: Biomedical Engineering, Cognitive Science, Computer Science, Electrical and Computer Engineering, Applied Mathematics & Statistics, Psychological and Brain Sciences.

reference

Institution : Johns Hopkins University
Direction : The Center for Language and Speech Processing
Website : www.clsp.jhu.edu
Disciplines : language modeling, natural language processing, neural auditory processing, acoustic processing, optimality theory, and language acquisition.
YouTube Channel : www.youtube.com/JohnsHopkins
Address : 3400 North Charles Street, Baltimore, MD, USA [ map ]
Information for applicants : goo.gl/mQuyY
Contacts : clsp@clsp.jhu.edu, +1 443- 997-6688 (tel.)

Carnegie Mellon University
The Human-Computer Interaction Institute (HCII)

“The mission of HCII is to understand and create a harmonious technology that enhances a person’s capabilities, his intentions and improve his social space through interdisciplinary research and education in design, computer and social sciences.”

Since 1985, HCII has been offering research and educational programs covering the full cycle of knowledge. It includes studies of social activity (work, play, communication) and social structures; design, creation and evaluation of technologies and tools to support social activities.

Main research interests and areas of HCII:

• User interface software.
• Cognitive patterns.
• Speech recognition.
• Understanding natural language.
• Computer graphics.
• Gesture recognition.
• Data visualization, visual design, multimedia.
• Computer support for collaboration.
• Computer music and theatrical skills.
• Social technology.

HCII is recruiting for undergraduate , postgraduate and PhD degrees .

reference

Institution : Carnegie Mellon University
Direction : The Human-Computer Interaction Institute (HCII)
Website : www.hcii.cmu.edu
Disciplines : user-interface software tools, cognitive models, speech recognition, natural language understanding, computer graphics, gesture recognition, data visualization, intelligent agents, visual interface design, multimedia, computer-supported cooperative work, computer music and drama, intelligent tutors, technical writing, and the organizational and social impact of technology.
YouTube Channel : www.youtube.com/CarnegieMellonU
Address : 5000 Forbes Avenue, Pittsburgh PA 15213-3891, USA [ map ]
Information for applicants : goo.gl/AqW92
Contacts : www.hcii.cmu.edu/contact-us , hcii@cs.cmu.edu

The educational market for speech technologies in Russia

The history of speech technologies (technologies, and not just scientific linguistics) originates from the upheavals associated with the organization of the Institute of Cybernetics in the USSR in 1959, the success story of which dramatically turned out to be the history of the failure and loss of world championship in this direction. The creation of the Institute of Cybernetics was partly due to Western successes, in particular, the demonstration on January 7, 1954 at the IBM New York office of the machine translation system (IBM-701).
Technologies of machine translation, text decryption, pattern recognition in the 50-60s. were brought to the level of the space program and the defense industry in the USSR and were supposed to prove the leading positions of the Soviet Union in the field of artificial intelligence modeling and computer design. The heyday of scientific thought at this time was associated with such surnames as N.D. Andreev, Yu.D. Apresyan, I.A. Melchuk, A.K. Zholkovsky, O.S. Kulagina, A.I. Berg, A.A. Lyapunov, M.L. Tsetlin, V. A. Uspensky, S. K. Shaumyan, and others.
In the 70s, the emerging approach to new frontiers in the field of artificial intelligence, speech recognition and synthesis, was for various reasons completely decentralized and, one might say, suspended in The 80s, when scientists were forced to switch from state funding to a grant basis.
By the end of the 80s and the beginning of the 90s. one can attribute the first attempts at the independent survival of individual linguistic schools and traditions, which later embodied their knowledge into commercially successful products and at a new stage in the development of speech technologies that realize their educational ambitions. About two of them - in our brief review.

Moscow Institute of Physics and Technology, ABBYY
Image Recognition and Text Processing

"Our goal is to make at FIVT (Faculty of Innovation and High Technology) the best teaching of Computer Science in Russia."

Since 2006, about fifty people have entered the department. Upon graduation, a job at ABBYY is provided, but graduates are not bound by any obligations to the company.

Main research interests and areas of RIOT ABBYY:

• Software Engineering.
• The basics of creating graphical user interfaces.
• The architecture of modern computers and operating systems.
• Development of distributed and client-server applications.
• Algorithms and data structures.
• Intelligent systems.
• Artificial Intelligence.
• User interaction design.
• Compilation theory.
• Logic and modeling reasoning.
• Design and analysis of algorithms.
• Linguistic basics of automatic word processing.

Students are admitted to the department starting from the third year of study (undergraduate, graduate).

reference

Institution : Moscow Institute of Physics and Technology, ABBYY
Faculty : Faculty of Innovation and High Technologies
Department : Image Recognition and Text Processing
Website : www.abbyy.ru/kafedra
Disciplines : design and analysis of algorithms, automatic word processing, applied theory of lattices, development of graphic user interfaces, intelligent systems, image recognition and processing, behavior modeling, perception and thinking, architecture development, client-server applications.
YouTube Channel : www.youtube.com/ABBYYVIDEOS
Address : Moscow, Klimentovsky per., 1, p. 18 [ on the map ]
Information for applicants : goo.gl/pA7x9
Contacts : (495) 408-4318, (495) 408-4633; fivt.fizteh.ru; upr@mail.mipt.ru, krivtsov@mail.mipt.ru.

St. Petersburg State University of Information Technologies, Mechanics and Optics (ITMO), Center for Speech Technologies

Speech Information Systems (RIS)

“We create products and technologies that help people understand others and be understood, making life in the global information community more efficient and safer.”

The Department of Voice Information Systems (RIS), opened in 2011, became part of the Faculty of Information Technology and ITMO Programming. The department trains specialists who are able to participate in research and design work in the field of speech information technologies with a specialization in the areas of speech recognition and synthesis, voice recognition, multimodal biometrics, in the field of design and development of information systems and software.

Main scientific interests and areas of RIS:

• Digital Voice Processing
• Speech recognition and synthesis
• Speaker Recognition
• Artificial Intelligence
• Multimodal Biometry
• Organization of software design and development
• Multithreaded programming
• Flexible software development models
• Information Systems Design
• System analysis and modeling of information processes and systems

Students with a bachelor’s or specialist’s diploma (preferably in the areas of information technology and programming) with general mathematical training are admitted to the department.

reference

Institution : St. Petersburg State University of Information Technologies, Mechanics and Optics (ITMO), Center for Speech Technologies
Faculty : Faculty of Information Technologies and Programming
Department : Speech Information Systems (RIS)
Website : www.speechpro.ru/career/learn-itmo
Disciplines : speech recognition and synthesis, voice recognition, multimodal biometrics.
Address : St. Petersburg, st. Krasutsky, 4 [ on the map ]
Information for applicants : May 17, 2011 - open day (registration ris@speechpro.com).
Contacts : +7 911 2643973; (812) 325-88-48; ris@speechpro.com