SPEAKER RECOGINITION

Name of presenters-Reuben Minael &Benjamin Mkwese
Name of Representer- Venance Luhemeja
Name of the Lecture Prof: Matti Tedre.
Lesson –Research field of computer science.

Introduction of Speaker Recognitions
The term Speaker recognition as the narrator of an amplified sounds or voice so as to make detection of the appreciated or approved sound has been determine in most of we Human being because we are able to be acquainted with someone just by hearing him / her when talking.
By definition!
Speaker recognition is the automatically process of making out distinguish between untruthfully and identified spoken based on his or her sound / voice.
Also speaker recognition has been categorized into two classes as
(a) Speaker Verification
(b) Speaker Identification.

Speaker verification, this is the authentication of sound or voice spoken to make
Proof of the recognized speech
While
Speaker Identification is the detection of sound or voice spoken to make discovery of the recognized speech.

HOW THE SYSTEM OF SPEAKER RECOGNITION WORKS
Firstly after storing of the sound or voice to be recognized, the amplified sound signals are converted from analogy to digital signals and make comparison of both current and previous speech and finally is machine instructions as workout exercise.

SOME REASONS WHY USING SPEAKER RECOGNITION
(i) Security operations
(ii) Access control to restricted services
(iii) Military operations.

Speaker recognition is very significant method using in our daily life tricks due to the fact that it helps us to be secure with our properties and make privacy of our own jobs.
Also helps in our governance community to make peace through hidden information to avoid misinformation and misunderstanding among people and society at all.
And lastly but not the least when military armies attack their enemies helps in secure of their information from enemies.

WHERE SPEAKER RECOGNITION CAN BE USED?
According to our lesson we came to observed that speaker recognition can be used in three different aspects techniques as;
(a)Authentication;
Allows people using to identify themselves nothing else but verification of their voices and sound.
(b)Surveillance;
In security agencies i.e. means of collecting information for inspection and observations as well.
(c)Forensic;
Identification proving of the recorded voice to be recognized by the user.

PROBLEMS OF USING SPEAKER RECOGNITION
Although speaker recognition seems as the best way of making people’s privacy but has got many challenges and bellow are some of challenges;
(i)Determination of the acceptable units of voice / sound
-It difficult to make the same correctly speech of the recorded and that used when doing speech recognition.
(ii)Restriction of signal boundary
-Also is difficult to make limitation of signal units when prepare speaker recognition due to the fact that we human being had variety of speech and sound.
(iii)Variability of sound / voice of speech.
-Also it is to make sound / voice to be vary in different accurate lather than to be in restricted point due to sinusoidal frequency of speech

HOW TO MINIMIZE CHALLENGES?
Although is difficulty to overcome the whole problems of speaker recognition but it can be minimized through various ways as stated below;
(a) Utilize the precisely expressions compulsory.
-it will easy to minimize when we use exactly sound and voices pertaining to recorded one.
(b) Maintain the preserved voice pitch.
-Also we should be supposed to upload or retain the recognized speech.
(c) be conscious of the upbringing circumstances
-We are supposed to understand the recognized order of speech.

Termination;
Speaker recognition is the best way of keeping our privacy securely but don’t make trustfully on that because when despoiled it can led a superior quandary as problems to your either country or company or Government or Army and finally you can be seen as untruthfully person.

Refference
(i)Nolan, F. The Phonetic basis of Speaker Recognition (Cambridge University Press, Cambrige 1983).
(ii)Martin. B.T, Dixon.R.N(1979)Automatic Speech and Speaker Recognition.
(iii)Schroeder. R.M(1985)Speech and Speaker Recognition.
(iv)http://www.scholarpedia.org/article/Speaker_recognition

Prepared by venance.