|
TOC
The Online Digital Talking Book Technology with Web and Mobile Clients project
was designed by Microsoft Visual Studio .NET 2008 beta 2 which is freely
available to download for everyone at the time it was launched.
Since this project has already been organized the source code and DLLs in the
single folder for both development and deployment, it can be easily deployed by
making a copy of all source code and its complied DLLs to a web-server which
have to be Microsoft Internet Information Services (IIS) version 6 or 7. In
addition, it needs a copy of Digital Talking Books (DTBs) in the same
web-server. That is because security protection issue in Silverlight --the main
Rich Internet Applications in the project-- protects cross-domain downloading of
all media used in the application.
If a developer needs to make some changes or modify some parts of the
application, it is required to have a set of Microsoft Visual Studio 2008 either
fully Professional version, the Express version or Trial version. In addition,
in the development server, it is indispensable to install ASP.NET AJAX and its
control toolkit. Moreover, in case of modifying the panel of control buttons
which makes from Microsoft Silverlight 2.0, it is also needed to setup Microsoft
Silverlight 2.0 SDK.
It may not be necessary to have a set of Microsoft Visual Studio, if the change
is about the soundtrack modification, for example, changing some wording in the
welcome voice, changing content of the soundtrack to another language. This kind
of modification needs three different tools.
Firstly, an audio editing tool likes the NCH WavePad for changing the wav file
(daisynow\onlinePlayback\Soundtrack\char_att_claire_us_eng.wav). In this audio
file stores all of the English characters and all equations that appear at the
standard keyboard. Moreover, it is included the welcome voice, instruction voice
of using in all dialogs, and alert voice in the program.
Secondly, a text editing tool for changing the soundtrack detail file in the
soundtrack folder (daisynow\onlinePlayback\Soundtrack\sound.xml), the content in
this file includes an audio filename, starting position, and ending position of
each phrase. This soundtrack detail file was organized by using XML (Extensible
Markup Language (XML) syntax. Each phrase was separated by element and its
properties (an audio filename, starting position, and ending position) were
saved as attributes
Thirdly the MP3 audio compression tool, due to the fact that an audio file in
WAV format cannot be used in the web-based streaming. In addition, an audio file
in MP3 format cannot be edited as a wave graph. Therefore, the output wave file
from the audio editing tool is needed to be converted to MP3 format by using the
audio compression tool. In this project, it takes advantage of the most popular
open source in MP3 encoding named LAME. The encoding bit-rate of the soundtrack
file which is used in this project is 96 Kbps Mono.
The most important issue in soundtrack editing is the recorded voice. The
recorded voice can be a human voice or a machine voice (from a text-to-speech
application). It would be great to have a human voice recorded as the
soundtrack’s application. Unfortunately, in the time I developed this project,
there was no a person who have the great pronunciation. Therefore, I decided to
use the machine voice (AT&T Natural Voices) from AT&T labs instead. This voice
is freely available on the ATT&T Labs website at
http://www.research.att.com/~ttsweb/tts/demo.php [5].
To operate this project, the prepared host must have at least Microsoft Internet
Information Services (IIS) version 6 or 7 on top of the Microsoft Windows
operating system. In case of providing to the user at home, it needs a high
speed internet connection. The most concerning issue of Internet connection is
the upload link because this web-server will be used to share all media files
(in the format of Digital Talking Book) on the server. Therefore, the upload
link of Internet connection will be the most important thing to concern.
However, due to the delay of cross-continent Internet connection, it would be
great to have multiple servers in the different continents according to
geographical distance. In this case, it could use the content delivery service
from an Internet solution provider which already has multiple servers on each
continent.
The maintenance is the most significant issue of the Online Digital Talking Book
Technology with Web and Mobile Clients project because, in the web-server, the
application is using by multiple users at the same time. Moreover, each individual user is
reading in the different book. The hardest part is to maintain the application
to be always up and running. There are two parts of maintenance issues that will
be covered in this report. The first part is about the server requirements. The
second part is about the quality of the content (Digital Talking Books)
requirements.
The server requirements
This project was initiated and designed by concerning in the scalability issue.
Therefore, the applications that will be installed on the server can be operated
on multiple servers, as in the Figure 1, or on
a single server by installing all server applications at the same server. The
information in the next five items will determine as the server requirements for
running all applications of this project.
·
High performance server(s) with high processor
clock speed
·
Redundant hard-drives for storing Digital Talking
Books
·
A domain name and server hosting with the
high-speed uplink network
·
The Microsoft Windows Server 2003 and IIS 6.0 web
server installed
·
The Microsoft .NET Framework version 2.0 (x86)
installed
The content requirements
According to project design in the both Middle
Tier and Data Tier, this project was designed to be compatible with the
“Specifications for the Digital Talking Book ANSI/NISO Z39.86-2005”. These
specifications define the format and content of the electronic file set that
comprises a Digital Talking Book as the standard [1]. Therefore, the
content requirement of this project is the well-formed digital talking book that
is valid to the specification of the specification of the digital talking book
mentioned above.
Each file set of a digital talking book that will be uploaded to the
web-server needs to pack together in a single folder. However, all digital
talking books can be organized in the hierarchical structure of directory. For
instance, in the first level of directory, it could be set for the book
category. In the second level of directory, it could be set for the book
sub-category. In the third level of directory, it could be set to store the
folders that each folder contains an individual digital talking book.
TOC
|