--- Forwarded Message from Bill Koulopoulos <[log in to unmask]> ---

>From: Bill Koulopoulos <[log in to unmask]>
>To: Language Learning and Technology International Information Forum   
<[log in to unmask]>
>In-Reply-To: <[log in to unmask]>
>Subject: Re: #9021.4 (!) Speech recognition software
>Date: Tue, 16 Dec 2008 07:48:04 -0500
>References: <[log in to unmask]>

Greetings to you all,

I have used MacSpeech Dictate for the last six months and have been  
happy with it.  Although some claim accuracy of 90% and up, in my  
experience it was about 85% ( I do have an accent by the way).  Yet,  
it has made a difference in the amount of typing I do and I  would  
recommend it.  It's not perfect, but it's the best (for Mac) out there.

Bill




On Dec 15, 2008, at 5:05 PM, LLTI-Editor wrote:

> --- Forwarded Message from Mike Ledgerwood  
> <[log in to unmask]> ---
>
>> In-Reply-To: <[log in to unmask]>
>> References: <[log in to unmask]>
>> From: Mike Ledgerwood <[log in to unmask]>
>> Subject: Re: #9021.3 (!) Speech recognition software
>> Date: Thu, 11 Dec 2008 19:40:33 -0600
>> To: Language Learning and Technology International Information Forum
> <[log in to unmask]>
>
> I hope I'm not stating the obvious, but Dragon (etc.) was created to
> help native speakers of a language and not to help FL learners.  This
> software was set up to adapt itself to a speaker's voice and oral
> facility.  Thus for foreign language usage, a non-native speaker
> could have an awful target language accent that would force the
> software to "bend over backwards" trying to adapt itself to the
> speaker.  It would try to come up with any word at all even if the FL
> speaker's voice contained little that would be understood by a TL
> user.  So, I can see a lot of frustration for an FL learner with poor
> oral skills.  Even if the software succeeded in producing
> comprehensible output, it would then support a less than desired
> level of oral language for many (most?) FL learners below the level
> of Intermediate (High?).
>
> I think the point of FL software should be to train the speaker for
> better oral language production and not to train the software to
> support the speaker.....
>
> Also have to admit to being sorry I spent a lot of money last year on
> Dragon (PC version) trying to help my 6th grader "write" better.
> However, his high voice timbre, quick frustration, and strong New
> York accent (ask how he is doing now in Alabama!) meant the software
> was abandoned after several hours of work.  And this was with a
> slightly non-standard native speaker of English.....
>
> Best to all, Mike Ledgerwood.
>
> P.S. Just a reminder that the IALLT Conf. call for proposals is now
> past.  However, those with a last minute idea might still consider
> doing a proposal.  The Program Comm. might still consider it.  I
> would love to meet many of you who contribute to this list who have
> never been to IALLT or not in a long time.
>
> Mikle D. Ledgerwood, Ph.D.
> Chair of World Languages and Cultures
> Professor of French
> Samford University
> [log in to unmask]
>
> on leave from the State University of New York at Stony Brook,
> 2007-2009.
>
> IALLT President-Elect
>
>
> ***********************************************
> LLTI is a service of IALLT, the International Association for
> Language Learning (http://iallt.org/), and The Consortium for Language
> Teaching
> and Learning (http://www.languageconsortium.org/).
> Join IALLT at http://iallt.org.
> Otmar Foelsche, LLTI-Editor ([log in to unmask])
> ***********************************************
>