Speech Recognition: - Shetul Chothani
Uploaded by
Sunil PillaiSpeech Recognition: - Shetul Chothani
Uploaded by
Sunil PillaiSPEECH
RECOGNITION
-SHETUL CHOTHANI
- PARAS KANERIYA
1. INTRODUCTION
Speech is the most natural and common
way of communication between people.
It would seem only natural that
computer development would eventually
progress to the point where people
would want to extend the human-
computer interface to include speech.
1.1 SPEECH
RECOGNITION(SR)
Speech or voice recognition is the ability
of a machine or program to recognize and
carry out voice commands or take
dictation.
It is a process by which a computer
interprets audible input from a user and
converts this data into a usable form.
Thus, SR is a technology that enables a
computer to understand the spoken word.
2. THEORY OF SPEECH
RECOGNITION
There are mainly three stages involved in
speech recognition:
Preprocessing: - Preprocessing involves
taking the speech input and converting it
into something the computer can use.
Recognition: - During the recognition
stage, the computer must identify what
has been said.
Communication: - Finally, in the
communication stage, the computer acts
upon the translated input.
2.1 PREPROCESSING
The first stage of the speech recognition
process is preprocessing.
In this stage, the data is sent as input to
the system.
The amount of this data must be kept to a
minimum.
The inherent challenge in this is to remove
the "bad" data, such as noise, without
losing or distorting the critical data
needed to identify what has been said.
HEADPHONE
General Input Device For Speech Recognition
2.2 RECOGNITION
Once preprocessing is completed, the
input data moves to the recognition
stage, where the primary work involved
in speech recognition is accomplished.
There are two main approaches to
attacking the speech recognition
problems in the recognition stage : a
Knowledge-based approach and a
Data-based approach.
2.3 COMMUNICATION
The final stage in the speech
recognition process is the
communication stage.
In this stage, the software system acts
upon the voice input it has received
and translated.
Thus the whole process of SR is
completed.
Mark Lucente, an IBM researche
moving one of the virtual objects
displayed on the screen.
“Put that…..” “Over there…..”
3. APPLICATIONS OF
SPEECH RECOGNITION
SYSTEMS
There are several examples of
applications of speech recognition
because speech recognition itself has
applications in many fields.
Most well known are the applications
like Dictation /text entry, Voice -
command systems, Information
services, Education, Security systems
etc.
3.1 DICTATION /TEXT ENTRY
First tune your audio setup, then train the
software to recognize your voice and after
that you can work with it anywhere you
would use a mouse or keyboard for input.
Freespeech 2000 from Phillips, Voice
Xpress from the Belgian company Lernout
& Hauspie, ViaVoice from IBM and Dragon
NaturallySpeaking from Dragon Systems
are the various s/w available in the market.
3.2 VOICE - COMMAND
SYSTEMS
With Voice-Command systems a user can
give commands to the computer by
talking to it and then some actions are
performed.
In the future voice-command systems will
allow information and communication
anywhere, anytime and anyway the user
wants it - in the office, at home, in the car,
kitchen, design studio, college and so on.
3.3 SECURITY SYSTEMS
Every day terabytes of data containing
personal information are send over the
World Wide Web.
Orders are placed and money is transferred.
For the security of the above transactions,
the company VeriVoice, based in Princeton,
New Jersey developed a system that uses a
Netscape Plug-in and the microphone that's
installed on most multimedia computers to
perform speaker verification.
3.4 INFORMATION SERVICES
Speech Recognition can also be applied
for information services like
communication.
Presently, research is done for doing
speech recognition over the telephone.
Instead of dialing numbers the caller can
choose the direction of the dialog by
talking to an automated service system.
3.5 EDUCATION
The use of speech technology in
education offers great new possibilities.
Learning to read and speak foreign
languages, transcription of lectures,
assistance for people with learning
disabilities are applications already in use.
Software programs developed at IBM's
Thomas J. Watson Research Center helps
children to learn to read.
4.1 CONCLUSION
In the recent years, voice and speech
recognition systems are becoming more and
more common, as evidenced by Sprint's
voice-activated telephone dialing system,
voice-controlled Windows applications, and
speech recognition devices in many
commercial industries.
The market for products continues to grow,
as people discover more and more
applications where they could be useful.
4.2 CONCLUSION
For many people in the past few years,
speech recognition has moved from just
being a novelty to becoming an important
tool used in their everyday lives.
Speech recognition technology will enter
our daily lives within a few years, for
some it already has.
The potential of SR technology will
enable humans to interact naturally with
machines.
4.3 CONCLUSION
This emerging technology is the next phase
in a series of steps to convert the computer
into an interactive, humanlike mind.
Indeed, as this technology improves, people
will ask their refrigerator for milk and tell
their cars where to go.
Gradual integration will take place and
someday you'll notice yourself wondering
"Gee, was that a computer I was talking to
or was it a human operator?"
THANK
YOU
?
