[pandorabots-general] Voice recognition

[Stanley E. Honour]

Here is an additional news item of interest:

*************************

Singularity Institute Blog Launched

KurzweilAI.net May 29, 2007

*************************

The Singularity Institute for

Artificial Intelligence (SIAI) has

launched a blog covering research

and outreach updates, videos,

articles, papers, events, goals, and

AI and cognitive science news. SIAI

is a not-for-profit research and

educational institute that fosters

scientific research, education, and

outreach to increase the likelihood...

http://www.kurzweilai.net/email/newsRedirect.html?newsID=6838&m=33305

 

Neil, P.s. NOT A PIPE DREAM !!   ALL THIS STUFF EXISTS NOW, JUST NEEDS SWEAT EQUITY.

ALSO THE ACCURACY OF VR HAS DRASITCLY IMPROVED IN THE LAST 5 YEARS,  NOW DRAGON NATURALLY SPEAKING IS BOTH VERY ACCURATE AND FAST WITH ALMOST NO TRAINING, FASTER COMPUTERS HAS NOT HURT EITHER;  REAL TIME RESPONSIVENESS STILL REQUIRES A HI END MACHINE WHEN YOU ADD THE TTS AND AVATAR OUTPUTS (WHICH MUST SYNC).

 

________________________________

From: pandorabots-general-bounces+stan=adnamis.org at list.pandorabots.com [mailto:pandorabots-general-bounces+stan=adnamis.org at list.pandorabots.com] On Behalf Of Stanley E. Honour
Sent: Tuesday, June 05, 2007 3:46 PM
To: pandorabots-general at list.pandorabots.com
Cc: mike Lc
Subject: Re: [pandorabots-general] Voice recognition
Importance: High

 

            Fred, Stewart here could be your first customer.  Fred here is a VR developer looking to setup a discussion group.

See responses below.   Current examples of TTS (text to speech) driving avatar 'actorbots' can be seen at www.haptec.com <http://www.haptec.com/>  and 'George' at jabberwacky.com these are chatbots linked to animation / flash systems.   The developer for George is Televirtual.com, an English company.  Both are proprietary.  Nuance.com can provide not just VR but TTS in various languages also but you must pay $$$$.  For example $750 for one copy of Japanese TTS (speaks English too with a cute accent)   There are pieces and parts that are shareware, but no one that I know of has done the necessary system integration to create FREE  Audio --> VR  -->  text --> chatbot I/O  --> text --> TTS  -->   sound out

                                                                                                + Parameters   -->  Avatar / Flash out.

P.S.  I do not personally know anything about 'Fred'.  I just happens these two emails passed my inbox within a few days of each other.  Proceed with due caution.   I also recall some A.I. researcher has already created a body language & sign language visual interface.   The Japanese are working hard on all of this.

 

From: Stan 

Glad to hear you are taking up the cause.   I am an IT manager now,  B.A. in experimental psychology.  A.I. news is a hobby of mine.   One goal of mine is to move chatbot tech to the next level; fully voice interactive.  I'm interested in emotion analysis in particular as applied to voice stream data.   This data is vital for a more friendly and compelling user interface and also for creation of actor bot avatar characters.   One aspect of basic human communication is a principle called "Reciprocity"; which means people instinctively seek to be on the same emotional plane, or wavelength.   A convincing and compelling A.I. character MUST be aware of the user's emotional state.   Much research has already been accomplished; all that remains is to put the pieces together.   Some software is already in labs but has not been applied to avatar control and text to speech + intonation.

For example there is the Japanese robot Replyee II, and Rollo Carpenter of Jabberwacky.com is working on such an interface.  (See George avatar) The main thing I can provide is references and examples of research I have collected over the past two years. 

By the way, you want people to provide personal info, but you have not identified yourself.  Most responses will be cautious.  

 

From: "Fred"

Ladies/Gentlemen;

 

This would be a huge undertaking to create an actionable product within the voice technology space, but where there is a will there is a way. It would require organization and support, and if we could facilitate this we could then create something tangible.

 

I would be willing to help put this organization together, but first we would need to gather if we have the talent and the willingness here to proceed. 

 

Please reply to this email address with:

your background info (resum¨¨, studies, expertise, passions, etc); availability; how you would like to participate; and requirements for participation.

 

We could then move foreword and we will attempt to put together a working group of individuals and in a month or so we will get together and collectively decide how to proceed.

 

Respectfully,

-Fred

 

 

 

________________________________

From: pandorabots-general-bounces+stan=adnamis.org at list.pandorabots.com [mailto:pandorabots-general-bounces+stan=adnamis.org at list.pandorabots.com] On Behalf Of stewart neil
Sent: Monday, June 04, 2007 8:49 PM
To: pandorabots-general at list.pandorabots.com
Subject: [pandorabots-general] Voice recognition

 

Hi

 

Has anyone done anything with voice recognition software?  I teach ESL and I want animated characters to respond to kids English.  Pipe dream I guess!?

 

Regards

 

Neil

  

________________________________

º†{µAb¶YA²€iÀ—¿—·xÀ[·vÙúš?·xU¶YöÄÀ@l²€ô¦°KA²€?¶YY·w?²€?Yahoo! Mail <http://login.yahoo.com/config/mail?.intl=hk> ¶YA¿—N¿|¸[?¸ZH! 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.pandorabots.com/pipermail/pandorabots-general/attachments/20070605/101a384d/attachment-0001.html