Abstract –
As the
technology advances the human needs to grow towards comfort. KMAIL, A Dynamic
mail using voice logging in cloud computing environment. Integrating voice communication into business
processes can accelerate resolution time, reduce mistakes, and establish a full
audit-trail of the interactions. In this
project, we are proposing a dual way of accessing the mail using Speech
synthesis & a normal way in a cloud computing environment. We propose a
generic framework consisting of a VXML-based service specification language.
Service providers using the proposed service-oriented architecture can offer to
their customers a protocol-neutral Web Service interface, thus enabling the
deployment of a general and integrated cloud solution for mail.
Key words: KMAIL, VXML,
cloud computing, speech synthesis.
I.
INTRODUCTION
Today,
people spend most of their time interacting with computers. Life is running at
a microchip speed. Over the last decade, these electronic tiny minuscule signals have fundamentally
revolutionized the way we live. People are spending more hours per day with
machines than humans. They communicate through mails across the globe for
official and unofficial information. Email & web have become part of our
life. These mailing systems are developed by different organization with
different modules and different feasibilities. But fail to reach the users who
are physically disabled and senior citizens. And also fail to provide better
ease for the users. The frequent and prolonged computer sessions may pose
physical health risks such as visual strain, harmful effects of radiation, and
posture and skeletal problems.
This problem can be overcome to some extent by
enhancing the applications with speech synthesis and speech recognition
systems. Here we propose a KMAIL reader system which is a unique system
developed for members of the system can exchange information and is also
supported by a reader which would read the contents of the mail by the
automatic speech recognition.
The cloud computing database system is used for
this system, in which we can enable Automatic Speech Recognition (ASR). The
selected mail in the inbox is processed to identify the name of the sender and
recipient, subject and the message. This text is normalized and segmented at
syllable level. Its corresponding syllable units are selected based on the
context and selected units are concatenated to get the wave file for the given
text.
It provides the user to register him as a
valid user of the KMAIL service and then allows the user to access the modules
of the system where the valid user can send mails, delete mails and receive
mails in the users language. The mails which the user receives are converted
from text to speech by using concatenative speech synthesis approach and an
appropriate wave file is generated and then the speech of the text is played.
The paper is organized as follows. Section 2
gives the details of KMAIL architecture. Section 3 gives the details of the
TTTS system architecture and different modules in the system. The conclusion in
section 4 followed by future scope in section 5.