Inventing for Humanity
We create, develop, and produce solutions to help people.
Interphraser&trade
Patent pending on system architecture and application
(Interlanguage Communication with Verification) #US12/214,284
Ingenuitor's Interphraser is a point-of-contact language solution that was initially developed to provide healthcare professionals an electronic handheld wireless tool to communicate immediately and effectively with Limited English Proficiency (LEP) and Deaf patients at a fraction of the cost of human interpreters, telephonic interpreters and video interpretation networks. The Interphraser solution will never completely replace live interpreters, but can be quick, error-free, and cost effective in situations that require rapid communication. The Interphraser is geared to work easily in a range of defined settings that could include: reconstruction, medical, border control, policing, and humanitarian aid.
1. The Problem
As extensively documented by NATO, the United Nations, and many other agencies, language barriers are becoming more and more of a problem. Over the last several years different solutions have been developed, used, and tested; however, there is no one solution that works for every situation.
Communication strategies include human solutions (ranging from the trained in-person interpreter to telephone or video interpretation all the way to the use of family or friends as untrained ad-hoc interpreters), electronic solutions (ranging from machine translations with speech recognition to an electronic travel aid), and printed material (ranging from multilingual phrasebooks with words to iconic language representations with only pictures and symbols). Just as a master artisan uses the right tool for the job, each solution has its place on the language tool belt.
The correct solution for communicating across the language barrier varies enormously depending on the situation. Translation of printed material is more than adequate in some situations. People—interpreters—are essential in others. But, trained and qualified interpreters, in person, by telephone, or video network are not always available or cost too much for the situation. Even if cost were not the issue, very few interpreters will be willing to work in a hazardous area at 3:00 in the morning, video interpreting may not be appropriate during a gynecological examination, and telephone interpreting does not work in areas where there is no phone reception, but each solution has its place.
The Interphraser system that we have developed is a complement to existing language solutions that will allow professionals to communicate in an easy-to use, accurate way with an encrypted, tamper-resistant, tamper-evident, and verifiable digital record of each encounter.
2. The Solution
Ingenuitor was incorporated in 2005 to fill a gap that we recognized in solutions to attack the growing language barrier. There were low-cost written solutions such as our best-selling "Pocket Medical Series" of books (>97,000 copies) on the market and there were high-cost human solutions (in-person, telephonic, and video interpreters), but there was very little in the middle. We realized that while machine translation was advancing dramatically with the use of sophisticated statistical processing, enlarged databases, advanced syntactical algorithms, and other techniques, the systems were not fast enough nor accurate enough for use in a medical setting. In addition, a "gisting" or "paraphrased" translation is not good enough when somebody's life is on the line or when diplomacy is critical. Science fictionesque devices such as the Universal Translator from Star Trek or the robot, C-3PO, in Star Wars that was fluent in over six million forms of human communication, are great in theory, but many years from seamless application due to errors in translation and/or speech recognition.
There were also commercial-off-the-shelf (COTS) type of travel aids and electronic phrasebooks on the market, but they only had a few medical phrases. The one or two devices on the market at that time that did cover the medical field, however, still had no way to record a patient's answer, no way to switch between languages, no way to record the interview history nor verify the accuracy of the data generated, no way to upload the interview history to a patient's file, etc. In response to the opportunity, we developed the Interphraser system to fill the gap, and which is positioned as an effective complement to existing language systems.
The Interphraser system is made up at least two units that form a communication loop. Each
unit has a processor, memory, screen, audio output system, and the technology to communicate
with the other unit. The communication system between the units can be hard-wired or wireless
(Bluetooth, 802.11, or other system). Obviously these specifications describe practically any
electronic device on the market today, and, in fact; the Interphraser system is programmed in
Linux and C++ with QT software modules that are very widely implemented. The content and
software can be installed and used on many kinds of units, and the two units do not have to be the
same. Practically that means that there are an infinite number of possible combinations.
One could have two equal units (one for the clinician and one for the patient), a "regular" unit for a doctor and a ruggedized larger unit for the patient, or two large tablets fixed permanently onto a cart in an "A-shape."
One deployment shown is on two Nokia N800 Internet tablets that communicate wirelessly via 802.11 (one for the doctor and one for the patient).
The Interphraser database has 1900+ medically related phrases in each of ten languages (English, Spanish, Russian, Korean, Farsi, Tagalog, Vietnamese, Mandarin, Cantonese, and Armenian) for a total of over 19,000 phrases. To guarantee the quality of the translations in the Interphraser, each phrase has been translated by a highly qualified professional translator, checked by an editor, and then proofread. The final phrase ends up being checked several more times as it is recorded and typeset. Problematic phrases are also back translated. It is as accurate as humans can make it.
In use an English-speaking doctor with a patient that speaks another language (e.g. Korean) will choose a category such as "pain." A list of phrases will appear that have pain as a keyword. The doctor can select one of those phrases such as, "Do you have chest pain?", and a signal will be sent to the patient's unit to display that phrase in Korean and to play a recording of a native speaker saying that phrase. If the phrase is a question, then a set of possible answers will also be displayed on the patient's screen. If the patient can respond, then they can select one of the possible answers. That answer will then be sent back to the doctor's unit and displayed in the doctor's language.
The Interphraser keeps a record of every phrase asked and answered, date and time-stamped that can be printed out or uploaded to the patient's file. If the patient cannot respond by touching their own screen, then the doctor can respond for them by selecting the answer on the doctor's unit. The history will indicate which unit provided the response.
In addition, every phrase that is displayed has an associated cryptographically calculated checksum that is also stored in the history. If someone attempts to change the record, they will not be able to match the checksum. The Interphraser system is thus tamper-resistant and tamper-evident. In other words, it is hard to change the interview history, and it is obvious when someone attempts to change that history.
The Interphraser uses a unique verification system for every phrase displayed, which is patent pending and that ensures that the phrase selected to be displayed on the patient's screen is what the doctor intended and is self-reinforcing, self-validating, and self-repairing. Similar verification and self-correction occurs when the patient's answer is sent back to the doctor's unit. Bear in mind that the Interphraser is not just a "translation" solution. It is a communicator that can also go between any combination of languages loaded, so it will work just as well going from English to Urdu, or Dari to Dari (for the hard of hearing or deaf), or even German to Arabic. The Interphraser software can therefore work in any country with people speaking any combination of languages.
The Interphraser system is also not limited to a particular phrase-set; using the same approach one could create literally thousands of different phrase-set used with thousands of different languages. The possible combinations are enormous, and the possible benefits to society are limited only by the imagination.
Because the Interphraser&trade will work with any language pair and with any set of phrases; it can help people anywhere. For example:
Each "cube" is a unique combination of the three variables: source language, target language, and phrase set. |
 |