TechTalks Event
Voice Recognition Technologies: Is It Time?
with guest expert John Morris of Drexel University
March 21, 2002
Audio
• Streaming
MP3
• Download
MP3 (Download
Tips)
We already use VR when we get an automated phone answering system that asks us to speak commands. What about speech to text and text to speech? What about authentication by voice? How fast can systems "learn" your voice? What's available commercially and what might campus applications look like in the near future? You missed the live event on March 21, but you can still get the straight scoop in this archive!
Guest Expert
John
Morris is the Coordinator of Academic Technology for the Department
of Information Resources and Technology (the infrastructure arm
of Drexel University). After 20+ years as a Professor of Architecture
and Architectural Engineering at various universities including
Tulane, University of Kansas, University of Michigan, and Drexel
University, he left his position as the Director of the Architectural
Engineering Program at Drexel to champion and institutionalize web-based
education and technology use in the classroom. His focus is on the
appropriate use of technology in education and the development of
"Best Practices" in Web-based learning including both pedagogical
and administrative issues.
Co-Hosts
Howard Strauss (above, left), Manager of Academic Applications
at Princeton University, is TechTalk's Technology Anchor.
Judith Boettcher is CREN's Executive Director.
Together, Judith and Howard will ask the really tough questions—and
relay the questions you email to them at expert@cren.net.
Background & Resources
Commercial Speech Recognition is a detailed, comprehensive meta-resource that our guest for this event says is *the* single most useful resource on speech recognition.
Another valuable tip from our expert for this Tech Talk is that Andrea Electronics is a comprehensive source for the various hardware peripherals useful in voice recognition technology implementation.
And one of the best, if not the best, article we have found that discusses the current state of voice recognition technologies is Voice Recognition: Sound Technology in a recent Federal Computer Week.
Perhaps even better is Sorting the ABCs of Speech Recognition on News.Com. Very comprehensive and dated only March 15, 2002!
Here's a good speech recognition meta-site with lots of links and resources.
Got an interest in doing speech recognition on Linux? This is your link.
The most useful consumer-oriented (user-friendly) website we found on voice recognition technologies (and it's comprehensive also) is Computing Out Loud.
What is the "Voice Web? This brief editorial discusses what it is not.
This is a concise guide to some terminology and potential uses of text-to-speech.
You may find some useful information in the archives of the W3C's "www-voice" public discussion list.
And here's a working draft of the WC3's Speech Recognition Grammar Specification for the W3C Speech Interface Framework. It's dense but very useful to skim as background for some of the deeper issues.
Voice XML - VXML - a lot of what you need to know about this area of research and implementation can be found at VXML Home.
Another locus of VXML info is VoiceXML Planet.
The Voice XML Review is useful for understanding where some of the current leading edge applications are being used, and how.
Speaking of Voice Recognition is a fairly Wired News article of interest. Its focus is on the Speech Application Language Tags (SALT) Forum.
Everything you want to know about SALT can be found on or linked to the SALTforum.
Another Wired News article, Voice Recognition: Still Trying, is a bit outdated but still illustrative of some common issues many people have with speech-to-text programs.
We seem to have "distributed" everything else, so why not Distributed Speech Recognition (DSR)?
Voice Recognition and Document Management in Education is a potentially valuable resource intended to simplify the process of selecting appropriate tools for administrators, teachers, and students.
The site is uninformative about the details, but York University apparently has a functioning voice recognition telephone directory that includes people's and departments' name recognition.
So does Swarthmore College - this student newspaper article details complaints about their system.
The "Speech at CMU Web Page" from Carnegie Mellon is an eye- (or ear-) opening experience.
Using voice recognition software in adaptive technology ways is a major driver of implementation on campuses. In its usual inimitable way, the University of Michigan's Adaptive Technology Computing Site has some good information on various programs. Worth browsing through.
The University of Toronto's Adaptive Technology Resource Centre, also has a very good set of voice recognition resources.
The National Center to Improve Practice in Special Education Through Technology, Media and Materials maintains a practical website with a k-12 focus, but k-12 may well be ahead of higher education in implementation.
Speaking to Write is yet another fairly comprehensive resource "which explores the use of speech recognition technology by secondary students with disabilities."
Authentication is a potentially valuable area of voice recognition. Maybe we're not using it in higher education much - yet - but check out this story about k-12 applications, including in the lunch line!
We guess that's just a niche application of voice recognition in customer relationship management. Here's a brief report on another, with potential on campus
And here's another product that says it puts a "fingerprint" stamp on voice recognition.
The Biometrics Institute provides links to many resources, plus definitions of terms such as "speaker verification" and "speaker recognition" - potentially useful to know in discussion about voice recognition.
�
Drexel's a pretty "wired" place and John's had a hand in much of what its done. Until we collect a more pertinent set of links to resources directly on voice recognitition technologies, here are some John/Drexel-related links to peruse.
Just for fun, here's a CNN report on Drexel's wireless initiative.
John also teaches. Here's the syllabus for CIVE 464 - Acoustics.
And here's a slide show from John entitled Online Course Development: From an Art to a Craft.
