Face-Tracking enhanced Speech-Recognition

Posted by admin on Jul 22nd, 2010 | 0 comments home title

This demonstration shows an innovative solution to a classic problem: how computer speech recognition can determine whether the user is talking to the computer, or to someone else. Face-tracking software processes video from a camera atop the PC screen to determine whether the user is facing the PC, or looking away. While facing the PC, audio-input and speech-recognition is enabled; while looking away, it’s muted. This technique also helps provide coarse end-pointing in a noisy environment. This demo highlights the use of two compute-intensive and highly-parallelizable workloads: image-processing and speech-recognition, and is currently only feasible on multi-core PCs. It’s also best-suited as a client-based workload (rather than cloud/server-based), due to video bandwidth.To learn more about the research that Intel is involved with please visit: www.intel.com/research

Want to Live Extra Years?

Mr. Computer on Speech Reco

Posted by admin on Jul 21st, 2010 | 2 comments home title

Mr. Computer, a Zen Computing Master, demonstrates speech recognition on his hybrid laptop as Stormi Weathers looks on.
Video Rating: 1 / 5

Patricia K. Kuhl is a Professor of Speech and Hearing Sciences and co-director of the Institute for Brain and Learning Sciences at the University of Washington. She specializes in language acquisition and the neural bases of language, and she has also conducted research on language development in autism and computer speech recognition. Kuhl currently serves as an associate editor for the journals Journal of the Acoustical Society of America, Neuroscience, and Developmental Science.
Video Rating: 5 / 5

Want to Live Extra Years?

Talk to Your Computer: Speech Recognition Made Easy

Posted by admin on Jul 15th, 2010 | 2 comments home title

Talk to Your Computer: Speech Recognition Made Easy

Talk to Your Computer shows how to make speech software work quickly and accurately. If you use IBM, L&H, Dragon, or Philips speech software, that book will save you hours of frustration and help you work in the most efficient way possible. The book describes how to tune up your computer for best accuracy, how to troubleshoot problems with your software, and how to matain an efficient work flow when dictating. A chapter on healthy computing guides you through setting up your workstation p

Rating: (out of 2 reviews)

List Price: $ 14.95

Price: $ 6.50

Want to Live Extra Years?

Quick Translation

Posted by admin on Jun 16th, 2010 | 0 comments home title

Quick Translation

Everybody in the translation business can tell you about customers who want translations produced very quickly. The stories are endless.

Most translators can type anything from 2500 words to 4000 words in a day, but often that is too slow to satisfy the requirements of a customer faced with an emergency situation.

Often a translation agency will split large translations which are required urgently between several translators, but in this case the problem of harmonising vocabulary can become quite difficult to solve and this is where computer translation memories come into their own.

For an experienced translator, the number of words he or she can produce per day can be considerably increased by using dictation. With the advent of small hand-held digital recorders with computer access, it becomes possible to send a dictation to a typist by email, have the typist type the translation in Word and then receive it back again by email. Of course, when the typed translation is received back from the typist, it has to be carefully proof-read and corrected where necessary.

Personally, I can speak much faster than I can write so I can send a dictation of 5000 words to one typist, send the second 5000 to a second typist, and so on. This makes it possible to translate well over 10,000 words per day.

Another way to produce translations very quickly is to use one of the speech recognition programs which have seen a lot of improvement over the last few years. Using one of these systems, the translator speaks into a microphone and the words appear on the computer screen. Some correction is always needed, but the system can work just as fast as one can speak.

There is another way of approaching the typing problem, which is to dictate to one of the computer speech recognition programs, where the words are typed by the computer directly on the screen. These programs have seen a great deal of improvement recently.

This dictation process becomes easier with experience and it does offer real savings in time.

And after all, it’s our time that we are selling in the translation business

More “computer Speech” Articles

Want to Live Extra Years?