i recently started to worry about eos not offering voice recognition, and i understood some developers consider this is not important, or even this would be “evil”. (this is not my point of view ;=) AND i am far of being the shadow of a developper, obviously far away of understandig 90% of the nowadays programming stuff)
So, voice recognition has to be used by now, with online services in most cases, but i remembered that in the late 2000/2001 i was using a program called “dragon naturally speaking” running on a desktop computer with very old technology compared to now, without the need of any connexion, and this program had a very accurate recognition of my voice after some initial training.
(i used it to dictate my courrier or do some technical writings)
so how is it possible that voice recognition seems to be so difficult to achieve without a very heavy technology and without beeing online 20 years later ??
this is a big mystery for me, and i suggest that this “old” technology could be ported to android (we have processors that are 100 or more times more powerfull than in those days, 8 cores, gigs of fast memory, etc etc)
the bad new is : i am totally unable to imagine how to get an old version of this program, and in addition i have not the beginning of an idea of how this could be ported to android devices…
Maybe one day we will start being creative again ??? (we, “humans”)
In general terms I agree, although it’s worth remembering that Dragon was (and is) a really big application taking up a lot of space. But I work with disabled people and for some of them, there really is no substitute for voice recognition. Not a massive market though, it has to be said.
Funnily enough, where I work we do have a couple of quite old copies languishing in a drawer, now that we have a more current version. But the old versions - one forgets quite how weirdly one had to talk. Every word had to be pronounced properly. It’s a far cry from the keyboard microphone button of today, unfortunately, and it’s not clear to me that many people would use it.
hi
by the time i used it in 2001-2003 i was quite happy of the very good recognition, mostly for doing some letters end some tech writting. Of course it was far of being so easy (long time of learning) as for today’s apps, but the amount of memory and storage used at this time was ridiculous compared to what we have on our smartphones nowaday… (not speaking of processors power…)
Anyways most of people i know are using speech to text in mails, whatsapp, sms, etc on their phones whenever they have a relatively long communication to write down. i’m not speaking of “during the driving” or “in public areas” use, but a regular use at home, while the keyboard on the phone is’nt very easy to use
Thanks for taking some time to give me your point of view !
Have a nice day !!
