Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 4 Feb 2010 08:14:46 -0800
From:      Diego Montalvo <dmontalvo@gmail.com>
To:        Mark <redtick@sbcglobal.net>
Cc:        freebsd-questions@freebsd.org
Subject:   Re: Voice Recognition System for FreeBSD...
Message-ID:  <aefec1611002040814l3823616fnb411dc51fe85669f@mail.gmail.com>
In-Reply-To: <776014.7729.qm@web81202.mail.mud.yahoo.com>
References:  <aefec1611002040722w3e316a81jcaa9a92e0e86436@mail.gmail.com> <776014.7729.qm@web81202.mail.mud.yahoo.com>

next in thread | previous in thread | raw e-mail | index | archive | help
At this point just looking for a simple one word or maybe two word per
"session" speech to text....


On 4 February 2010 08:03, Mark <redtick@sbcglobal.net> wrote:
> --- On Thu, 2/4/10, Diego Montalvo <dmontalvo@gmail.com> wrote:
>
>> From: Diego Montalvo <dmontalvo@gmail.com>
>> Subject: Voice Recognition System for FreeBSD...
>> To: freebsd-questions@freebsd.org
>> Date: Thursday, February 4, 2010, 9:22 AM
>> Trying to test out voice recognition
>> software for FreeBSD is there a
>> recommended system (ports). Simply want to speak and have
>> the computer
>> "type" what ever it transcribes...
>>
>> Thanks in advance!
>> Diego
>> _______________________________________________
>> freebsd-questions@freebsd.org
>> mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
>> To unsubscribe, send any mail to "freebsd-questions-unsubscribe@freebsd.=
org"
>
> IBM had a speech recognition engine that was open source, but it looks li=
ke they pulled it off the open list. I think is was Via-Voice
>
> Look at audio/sphinx.
>
> I did some research sometime back, voice to text is the golden egg. =A0Ev=
en with Microsoft based software, every person using it must "train" the so=
ftware. Getting doctors and nurses to sit down and spend the time going thr=
ough a list of words that they must repeate 10 to 15 times.
> Then they tend to alter the way they speak toward the end. When normal in=
put is tried, too many errors. Transcribers must go over the text and audio=
 file. Not much of a money or time saver at this point.
>
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?aefec1611002040814l3823616fnb411dc51fe85669f>