Lamia_Al_Madi_352_-_HW_3

joinherbalistAI and Robotics

Nov 17, 2013 (3 years and 6 months ago)

56 views

Name: Lamia Al Madi

352
-

Assignment 3

ID: 207410006

Systems
Speech


:
systems

ependent
nd
I

aker
ependent and Spe
D

eaker
Sp

Spe
aker

dependent is
a
speech recognition software that is trained to


recognize its users' speeches. This
c
ould

be able

by

training the

software


to identify and understand. It has been said that
spe
aker

dependent

has


limitations
in terms of understanding speakers, so it understand
s

only


selected speakers.
(1)

These systems cost less and developed easily but


are
less flexible.
(2)

On t
he other hand,
spe
aker

independent system is also a speech


recognition software that can recognize a number of voices with no


training; however; it has limitations on the number of vocabulary.
(1)


It costs more and not easily developed.
(2)


gnition and Voice Recognition systems:
Speech Reco

Speech recognition system is when a speech device or engine (e.g. a


computer) recognizes human voices generally.
(1)

It takes orders by listening to human voices and conveys them into texts,


and it is known
for

the
ability to recognize different human voices


without an actual training. For example, voice dialing, voice routing and


data entries.
(3)


A
dvantages of

speech recognition
i
nclude
: it doesn’t have to be trained


to function
,

and it is faster.
A

disadvantage

of speech recognition is that


it has limited

vocabulary.
(4)


Voice recognition is a system
that restricts itself by recognizing the users'


unique voices, and it focuses

more on recognizing the voices of users


and their styles
more than the content itself.
(1)

For example, voice


recognition devises
for security access.

An advantage of voice recognition is that there will be no spelling errors,


but some disadvantages

a
re
:
it need
s

a huge memory space, and it



might
be difficult to work when the speaker is in a noisy place.
(5)


As we can see, speech recognition is the s
ame as spe
aker

independent


whereas
voice recognition is the same as spe
aker

dependent.

All these syst
ems are relevant to translation mainly because they include


the language. This language needs to be translated into another;


however; these systems concern themselves with the literal translation


of the speech more than just the content. These systems c
an be used to


break down language barriers and translate from one language to

another in a situation.

For example, when a native English is in Athens

airport, where the
average

of English speakers is very low, he can use

speech recognition system
s

to t
ranslate what he wants the receiver to

understand. This can be done mostly by using cell phones applications

and programs; however; there is a risk of mistranslating some

ambiguous words. We can say that the entire process is related to the


translation

field and its linguistic aspects.

Examples:




VozMe

is a text to speech system.

-

It is free.

-

It does not offer an Arabic language option (only 4 languages included).

-

Updated on 3
-
3
-
2008.

-

No specific users. All users can use it.




Talk to Me

is a speech to
speech

mobile application.


-

It is free.

to Spanish, French, UK, German

English
ly from
No Arabic option (on

-
and Italian).

-

V 2.0.0.

-

Probably English speakers.







iPhone

Voice recognition

is a Speech to
text

application

(alth
ough
through
this

study,
we can say that
this is a case of speech recognition
rather than voice recognition because the application is not trained to
particular voices

and can function with all voices
).


-

It is free.

-

It functions with all languages in
cluding Arabic (although Arabic users
face some technical problems with
the
Arabic recognition).


-

Version 4.

-

All language users.


References:



1
-

LumenVox Resource Center.
Speech Recognition Glossary
.
<
http://www.lumenvox.com/resources/tips/tipsGloss
ary.aspx
>.

2
-

What is Speech Recognition?

18
-
6.
<
http://www.speech.cs.cmu.edu/comp.speech/Section6/Q6.1.html
>.

3
-

Speech recognition
.

19
-
3
-
2011.
<
http://en.wikipedia.org/wiki/Speech_recognition
>.

4
-

Sugumaran, Kandasamy.
Speech Recognition Systems
.
<
http:/
/www.doc.ic.ac.uk/~nd/surprise_95/journal/vol1/ks4/article1.ht
ml
>.

5
-

Landmark College.
Voice Recognition
.
<
http://www.landmark.edu/institute/assistive_technology/voice_recog
nition.html
>.

6
-

Vozme.

<
http://vozme.com/index.php?lang=en
>.

7
-

Best Android.
(2009
-
2011).
Speech
-
to
-
Speech Translation with Talk to
Me.

<
http://www.bestandroidappsreview.com/2010/02/speech
-
to
-
speech
-
translation
-
with
-
talk
-
to
-
me.html
>.

8
-

Gadgets Reviews.
iPhones get speech to text app.
<
http://www.gadgetsnreviews.com/iphones
-
get
-
spee
ch
-
to
-
text
-
app/3414.html
>.