LLTI Archives

April 2001, Week 4

LLTI@LISTSERV.DARTMOUTH.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
LLTI-Editor <[log in to unmask]>
Reply To:
Language Learning and Technology International Information Forum <[log in to unmask]>
Date:
Wed, 25 Apr 2001 08:07:17 EDT
Content-Type:
text/plain
Parts/Attachments:
text/plain (65 lines)
--- Forwarded Message from Derek Roff <[log in to unmask]> ---

>Date: Tue, 24 Apr 2001 07:53:16 -0600
>From: Derek Roff <[log in to unmask]>
>To: LLTI-Editor <[log in to unmask]>, Gary Dauphin <[log in to unmask]>
>Subject: Re: #6078.2 Voice analysis / waveform
>In-Reply-To: <[log in to unmask]>

Gary-

Most of the speech to wave-form generators do nothing useful.  That
is, they display a "picture" of the sound, but this picture can not
be imitated by the student, and there are few tools provided to help
the student adjust their speech to more closely approximate the
model.   Using these ineffective tools just frustrates the students.

If you record two native speakers and display their speech
wave-forms, the pictures usually look entirely different.  Student
speech, will look different from either native speaker, whether the
student has pronunciation problems or not.  The problem is that the
wave-form generators accentuate and display unimportant differences
between individual speakers, rather than adherence or divergence from
the language norms for elements necessary for effective
communication.

Obviously, it is a hard problem to analyze a speech sample, and
discard the idiosyncratic information, while graphing the language
critical elements.  Transparent Language has upgraded their speech
sample graphing significantly for version 8.  I haven't tested it
enough to know how useful it is, but it is inexpensive and has an ESL
version.  Signalize, mentioned by Sara, has the reputation of being
the best and most useful speech analysis/graphing tool.

Derek

>> From: Sara Wilson <[log in to unmask]>
>> Subject: Re: #6078 Voice analysis / waveform
>
> Gary,
>
> I think the software you're referring to, or at least an
> equivalent,  is Signalize.  We used to use it here, and information
> about it can  be found at http://www.infosignal.com/index.htm
>
> Sara

>>> From: Gary Dauphin <[log in to unmask]>

>> A user at Auburn university is interested in a Macintosh
>> application that records a speakers voice, and draws out a
>> waveform graph on the screen.
>>
>> I remember years ago one of the LLTI-member schools had a program
>> that wrote in-house, I think using a Hypercard XCMD, that did just
>> that.
>>
>> This particular user is interested in it for ESL, but I don't
>> think the language really matters


Derek Roff
Language Learning Center, Ortega Hall Rm 129, University of New Mexico
Albuquerque, NM 87131  505/277-7368 fax 505/277-3885
Internet: [log in to unmask]

ATOM RSS1 RSS2