[Xerte] TTS (formerly Re: Xerte 2: MVC approach to displayed text)

Alistair McNaught Alistair.McNaught at HEAcademy.ac.uk
Tue May 3 14:14:50 BST 2011


There are some very good free text-to-speech tools.   A couple of
portable solutions that run directly from a memory stick without Admin
rights include Balabolka and DSpeech. The issue with both of these is
that they will simply take your default system voices which tend to be
golden oldies like Microsoft's Sam or Microsoft Mary. If you are lucky
enough to live in Scotland and are part of the Scottish education system
there is free voice Heather licenced to educational institutions. JISC
TechDis has been given the go-ahead to develop a similar high-quality
English voice but this is at an early stage so won't be ready for a
while. Zamzar.com does a good web-based conversion to MP3 as does
RoboBraille.org. The free version of naturalreaders.com also does a
reasonable voice but to save as an MP3 you need to upgrade to the
commercial version.

 

Paralink is very good though - thanks for the link - I'd not seen it
before.

 

Alistair

 

From: xerte-bounces at lists.nottingham.ac.uk
[mailto:xerte-bounces at lists.nottingham.ac.uk] On Behalf Of Kemp
Johnathan
Sent: 03 May 2011 14:00
To: Xerte discussion list
Subject: [Xerte] Re: Xerte 2: MVC approach to displayed text

 

I have looked at TTS software quite recently, but with a view to finding
a free solution.

 

There are a number of offerings that provide some kind of free download,
but in general the quality of the voices is poor. The software available
is able to provide high quality TTS but only if you buy the better
voices. This was why I was impressed with paralink.com's offering. It
was free to use and delivered a good quality speech rendering,
particularly the male voice.

 

I would prefer something that I could install locally as it is unlikely
to be significantly affected by a change in business model, but for the
time being paralink's offering is the best free offering I have come
across, particularly so once you realise that despite the web site not
offering the option to download an audio file of the reading, you can
still make your own recording directly to Audacity without the need for
a microphone by tinkering with your audio mixer settings.

 

I am afraid we have both wondered somewhat of the original topic of this
posting :-)

 

Kind regards

 

Johnathan

 

On 3 May 2011 13:27, Julian Tenney <Julian.Tenney at nottingham.ac.uk>
wrote:

Have you looked for some windows applications for generating TTS sound
files from text? I haven't for a while, but the more recent voices are
really good. There was an application - I forget what it was called now
- that allowed you to generate the mp3s directly on windows, using the
SAPI stuff on your system, so you could install whatever voices you
wanted. Must be something out there.

 

From: xerte-bounces at lists.nottingham.ac.uk
[mailto:xerte-bounces at lists.nottingham.ac.uk] On Behalf Of Kemp
Johnathan
Sent: 03 May 2011 13:02
To: Xerte discussion list
Subject: [Xerte] Xerte 2: MVC approach to displayed text

 

Recently someone posted for help in locating where some text strings
were contained in pages as they wanted to convert the page for use in a
language other than English.

 

Would it be a good idea to establish a standard by which all displayed
text in a Xerte Page model was entered via the page's wizard?

Currently pages like the quiz page have embedded in the code in the .rlm
file statements like 

fb = 'The quiz is complete.

tooltip="Next Question"

fb += '<br><br>You scored '

 

If all such text was replaced with data calls to an external xml source
then these strings could be readily modified using an xwd defined form.
This could open up Xerte more readily for use internationally.

 

I have been doing some testing recently with Page Wizards and have found
that I can add a language block to the xwd and xml files e.g.

 

xwd content

 

<wizard>

<!--DIALOG
INTERACTION-------------------------------------------------------------
------>
<dialog hint="An interaction comprising a sequence of text-entry
interactions forming a short dialog.">

<text type="TextArea" height="100" label="Text"/>
<attempts type="ComboBox" width="100" options="1,2,3,4" label="Attempts
before hint"/>

<narration type="media" label="Narration" optional="true"/>

<newNodes>
<dialogStep><![CDATA[<dialogStep name="question" question="Write your
question here." answer="Enter Answer" hint="Enter a
hint"/>]]></dialogStep>
</newNodes>

</dialog>
<dialogLang>
<tryAgain type="TextInput" label="Try Again Prompt"/>
<continue type="TextInput" label="Continue Prompt"/>
<showAnswer type="TextInput" label="Reveal Prompt"/> 
</dialogLang>

<dialogStep menuItem="Question" icon="icComment" remove="true" max="12">

<question type="TextArea" height="100" label="Question"/> 
<answer type="TextInput" label="Answer"/> 
<commentary type="TextArea" height="100" label="Commentary"/> 
<hint type="TextInput" label="Hint"/> 

</dialogStep>

</wizard>

 

XML content

 

<learningObject>
<dialogLang tryAgain="Try Again" continue="Press Enter to continue"
showAnswer="The correct answer has been revealed" />
<dialog linkID="" text="Enter text for the page here" attempts="2">
<dialogStep name="question" question="Write your question here. 1"
answer="Enter Answer 1" hint="Enter a hint 1" />
<dialogStep name="question" question="Write your question here. 2"
answer="Enter Answer 2" hint="Enter a hint 2" />
</dialog>
</learningObject>

 

Rlm examples

 

The model file can then access the language strings defined in the
wizard by calls to the xml data 

 

pageIcon.templateData.learningObject[0].dialogLang[0].showAnswer

 

(With forthcoming changes to the Page Wizards this will soon be reduced
to   pageIcon.templateData.dialogLang[0].showAnswer  )

 

I have tested this and it works.

 

This approach has the distinct advantage of providing a separate page in
the wizard form, for the language strings, thus avoiding having forms
that are significantly longer than the window in which they are
displayed. It also separates the display related text from the content
related text. 

 

However it requires separate blocks outside of the <dialog></dialog>
tags which may not prove compatible with pageTemplates. 

 

Can anyone more aufait with pageTemplates or On-line toolkits see any
problems with this approach in terms of it being transferred to
pageTemplates /  On-line toolkits, or can you think of a better way that
would be more compatible with the way the xml and xwd data is organised
in the pageTemplates environment?

 

As a result of other recent development work I have carried out on the
Page Wizards I believe a single .rlm file that could be used both within
Page Wizards and  pageTemplates (and presumably therefore On-line
toolkits) is now achievable. I would like therefore to consider the
implementation of better support for language support in a context that
considers both Page Wizards and pageTemplates based pages.

 

Kind regards

 

Johnathan

 

 

 

 

 

 

This message and any attachment are intended solely for the addressee
and may contain confidential information. If you have received this
message in error, please send it back to me, and immediately delete it.
Please do not use, copy or disclose the information contained in this
message or in any attachment. Any views or opinions expressed by the
author of this email do not necessarily reflect the views of the
University of Nottingham. 

This message has been checked for viruses but the contents of an
attachment may still contain software viruses which could damage your
computer system: you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation. 


_______________________________________________
Xerte mailing list
Xerte at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte

This message and any attachment are intended solely for the addressee
and may contain confidential information. If you have received this
message in error, please send it back to me, and immediately delete it.
Please do not use, copy or disclose the information contained in this
message or in any attachment.  Any views or opinions expressed by the
author of this email do not necessarily reflect the views of the
University of Nottingham.

This message has been checked for viruses but the contents of an
attachment
may still contain software viruses which could damage your computer
system:
you are advised to perform your own checks. Email communications with
the
University of Nottingham may be monitored as permitted by UK
legislation.



 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte/attachments/20110503/6c67036f/attachment-0001.html>


More information about the Xerte mailing list