I use the latest Dragon, and I've found it light-years ahead of
anything else I've ever tried. But I did need to "train" it with a
half hour or so of reading a speech, and it's not 100 per cent
accurate, especially with unusual words.
Based on my experience, here's a few of suggestions:
1
>I'd have to see it to believe it. Even if it got the sounds right 100%
>of the time (highly unlikely), there's simply no way it could deduce
>homonyms correctly, which would be needed for proper punctuation...
Bight yore tong!
I beg to differ. This software is able to distinguish among homonyms with a
high degree of accuracy, and as stated earlier, it learns through experience.
Worst case, someone might use it to repeat interviews into the computer - which
would still save a boatload of time and effort.
I have used
Photographer
Indiana Historical Society
450 W. Ohio St.
Indianapolis, IN 46202
(317) 232-4592
[EMAIL PROTECTED]
-Original Message-
From: Computer Guys Announcements and Discussion List
[mailto:[EMAIL PROTECTED] On Behalf Of Daniel Else
Sent: Thursday, August 02, 2007 10:34 AM
To: COM
I'd have to see it to believe it. Even if it got the sounds right 100%
of the time (highly unlikely), there's simply no way it could deduce
homonyms correctly, which would be needed for proper punctuation at
least.
I don't see any way of doing this project with any degree of accuracy
short of havi
MAIL PROTECTED]
-Original Message-
From: Computer Guys Announcements and Discussion List [mailto:[EMAIL PROTECTED]
On Behalf Of Daniel Else
Sent: Thursday, August 02, 2007 10:34 AM
To: COMPUTERGUYS-L@LISTSERV.AOL.COM
Subject: Re: [CGUYS] Speech recognition software
The latest ver
The latest version of Dragon NaturallySpeaking claims to not require "learning
the voice" before transcribing voice to text. It should therefore be perfect
for your use. I've used several versions of the software, and I find it
excellent (though you always must check for accuracy). It will proba
Here's the set-up: We have hundreds of oral histories on cassette tape. I've
successfully transferred some of these to CD using Audacity (freeware). We
have several people transcribing these histories so we have written copies. Is
there some speech recognition software we could use to for tr
