Converser for Healthcare for Interactive Speech-to-Speech and Text-to-Speech Translation

Spoken Translation will be showcasing the world’s first system for interactive speech-to-speech and text-to-speech translation at SpeechTek 2005, August 1st through August 4th, at the Marriott Marquis in New York City. STI’s first vertical application of its ConverserT system, Converser for Healthcare, allows people who don’t speak the same language to hold wide-ranging health-related conversations in real time, without a human interpreter.

Until now, there has been no simple, private, reliable, and inexpensive way to initiate a cross-lingual conversation 24/7. STI Founder and CEO Dr. Mark Seligman says, “Converser represents an exciting breakthrough for both the speech and translation industries: multimodal interpreting systems can improve communications not only for healthcare, but in customer service, business-to-business, education, diplomacy, government activities, personal relationships, and in many other areas.” Robert Aiudi, STI’s business development and sales leader, adds, “While Converser cannot and should not always replace human interpreters, the system does provide significant value propositions, including reduced interpreting costs, 24/7 accessibility to interpreting services, protection of privacy, transcript generation, and verifiability.”
 
What is an interactive speech-to-speech/multimodal translation system? It’s software (running on a Tablet PC, Laptop, etc.) in which the user speaks into a microphone in one language and then hears the computer pronounce a translated version in another language. The results have a high level of accuracy, since the user has been able to monitor and correct its progress at several stages.
 
Converser also enables communication through multiple modes of data entry: In addition to speech recognition, you can also use handwriting recognition technology, touch-screen keyboards, or standard keyboards. Behind the scenes, multiple technologies work together: speech recognition, machine translation, text-to-speech, and STI’s unique user interfaces are tightly integrated with proprietary word and phrase databases. The proprietary system automatically provides specially controlled re-translations, or “back-translations,” of input sentences, lending confidence that the translation says what you mean. And if speech recognition or translation errors are found, they can be easily corrected via unique Speech CuesT and Meaning CuesT technology.
 
 
Source: Spoken Translation, Inc.