17.1885, FYI: Artificial Speech Listening Tests

[linguist at LINGUISTLIST.ORG]

LINGUIST List: Vol-17-1885. Mon Jun 26 2006. ISSN: 1068 - 4875.

Subject: 17.1885, FYI: Artificial Speech Listening Tests

Moderators: Anthony Aristar, Wayne State U <aristar at linguistlist.org>
            Helen Aristar-Dry, Eastern Michigan U <hdry at linguistlist.org>
 
Reviews (reviews at linguistlist.org) 
        Laura Buszard-Welcher, U of California, Berkeley  
        Sheila Dooley, U of Arizona  
        Terry Langendoen, U of Arizona  

Homepage: http://linguistlist.org/

The LINGUIST List is funded by Eastern Michigan University, Wayne
State University, and donations from subscribers and publishers.

Editor for this issue: Svetlana Aksenova <svetlana at linguistlist.org>
================================================================  

To post to LINGUIST, use our convenient web form at
http://linguistlist.org/LL/posttolinguist.html.


===========================Directory==============================  

1)
Date: 23-Jun-2006
From: Tina Bennett < cbennett+linglist at cs.cmu.edu >
Subject: Artificial Speech Listening Tests 

	
-------------------------Message 1 ---------------------------------- 
Date: Mon, 26 Jun 2006 12:57:59
From: Tina Bennett < cbennett+linglist at cs.cmu.edu >
Subject: Artificial Speech Listening Tests 
 


The Blizzard Challenge 2006:
Evaluating corpus-based speech synthesis on common data
http://www.festvox.org/blizzard/

LISTENING TESTS:

In order to investigate corpus-based speech synthesis techniques we have
devised a large experiment in which research laboratories around the world
used the same speech databases to build synthetic voices.

Year two of the Blizzard Challenge is upon us, and once again, we would
like to have as many people take part as possible.  We also feel that
diverse bodies of participants (i.e. those who are knowledgeable about
speech, those knowledgeable about technology, and lay people) allow for
interesting comparisons.  

In the web-based listening tests, various waveform files are played to the
listener who then is asked to score them on a 1-5 scale or in some tests,
to type in what they hear.

To participate in the listening tests please go to:

http://www.speech.cs.cmu.edu/blizzard2006/register-R.html

You will be asked to register, then listen to and evaluate a set of
sentences.  It takes around 30-60 minutes to complete all the tests.  It is
best if the tests are completed in a single session, but it is possible to
leave and return later, continuing from where you left off.

When you have finished, please give us as much feedback (via the exit
questionnaire) as you are willing.  Native language information, etc. will
be especially useful.

Thanks for helping to improve the quality of spoken output!

Tina Bennett
Ph.D. student, Speech group
Language Technologies Institute
Carnegie Mellon University 



Linguistic Field(s): Not Applicable





 




-----------------------------------------------------------
LINGUIST List: Vol-17-1885