The boundaries and limitations on universal translation are known only to be no doubt beyond what we can imagine today. Certainly while there are exciting developments being made by Google, Baidu, and other organizations, extant technology has not yet begun to to be utilized on par to its potential. Perhaps there are sociological reasons for this inhibition in developments, but innovators are those who dream, and push their dreams onto the world of people through the demonstrable material. It’s not a matter of wanting people to ‘buy more gadgets’ but giving them the idea that sparks them to dream new dreams. Our innovations are all about human experiences; this is what the digital offers us, and by which it challenges us as well.
Specifically, I envision as a prototype a server. The server can be small, about the size of a pocket wifi unit, because microphones can be small and retain their sensitivity. Users may lagin to the translator via apps on hand held devices such as smart phones and pads, or laptop computers. International offices may also have their computers dedicated to the system; overall performance would be the same. The unit may be a standalone device, on installed into a computer’s existing hardware architecture. The unit’s software would be dedicated solely to human language translation, and contain very large databases to accommodate not only all of the world’s major languages but its dialects and accents as well. Further, it must be able to recognize these languages automatically without manual switching. The other database will be massive and present tremendous challenges: that will be the idiomatic translator database and it will need to accurately translate nuances in phrase and usage. Without a concentrated effort by organized software entities who remain first preoccupied with pleasing Wall Street before human progress, it could take decades for a whole language database to develop organically.
The unit will of course handle simultaneous translation in multiple languages and its software, like today’s language switchers, will require very little computation to complete pre-scripted language templates. It will send results via a micro speaker unit that clips invisibly and comfortable to each ear and amplifies the translated speech. The intonation and individual specifications and peculiarities of the speaker will be adjusted and reproduced by settings initiated in reference to database 3, the articulatory database. As the signal will be the decisive factor in sound production and the translations themselves will be for the most part muted by the micro ear phones, communication will be a pleasant experience with few distractions.