Normal view MARC view ISBD view

Command by speech recognition / Ardian Syah Mohd Yusof

By:

Ardian Syah Mohd Yusof

Material type: Text

TextPublication details: Kuantan, Pahang : UMP, 2008Description: 88 p. : ill. ; 30 cm. + 1 computer discISBN:

THE0007080(Local)

Other title:

Command by speech recognition [electronic resource]

Subject(s):

Dissertation note: Project paper (Bachelor of Electrical Engineering (Electronics)) -- Universiti Malaysia Pahang - 2008 Abstract: Speech recognition is a topic that very useful in many applications and enviroments in our daily life. Generally, speech recognizer is a machine which understand humans and their spoken word in some way and can act thereafter. In daily usage, for example, it can be used in a car environment to voice control non-critical operation such as dialing a phone number to ensure a maximum control to the car and enhance the safety. A different aspect of speech recognition is to facilitate for people with functional disability or other kinds of handicap. The system develop for this speech recognizer will be done using MATLAB and for this project, the system will recognize discrete word only and not a sentences or a robust speech. There are two main operations in this speech recognizer which are generating the voiceprint and store it as template in the word bank and recognizing the word spoken by comparing it with the template stored in word bank. The voiceprint is create by extracting it’s MEL Frequency Cepstral Coefficient (MFCC) which is the default number of coefficients need to be extracted are 12. To get more accurate result, 20 coefficients will be extracted. For recognizing purpose, Dynamic Time Warping (DTW) method is use. DTW will calculate the distance of two vectors which are the word spoken and the stored voiceprints and it will recognize the word as the same word if the the distance is the lowest or in other words, nearly zero.

Tags from this library: No tags from this library for this title. Log in to add tags.

Average rating: 0.0 (0 votes)

Holdings ( 2 )
Title notes ( 2 )

Holdings
Item type	Current library	Call number	Copy number	Status	Date due	Barcode
Final Year Report	UMPLIB PEKAN	TK7895.S65 A73 2008 rs Bc. (Browse shelf(Opens below))	1	Not for loan		0000037436
Final Year Report	UMPLIB PEKAN	CD 3282 \| TK7895.S65 A73 2008 rs Bc. (Browse shelf(Opens below))	1	Not for loan		0000037437

Project paper (Bachelor of Electrical Engineering (Electronics)) -- Universiti Malaysia Pahang - 2008

Speech recognition is a topic that very useful in many applications and enviroments in our daily life. Generally, speech recognizer is a machine which understand humans and their spoken word in some way and can act thereafter. In daily usage, for example, it can be used in a car environment to voice control non-critical operation such as dialing a phone number to ensure a maximum control to the car and enhance the safety. A different aspect of speech recognition is to facilitate for people with functional disability or other kinds of handicap. The system develop for this speech recognizer will be done using MATLAB and for this project, the system will recognize discrete word only and not a sentences or a robust speech. There are two main operations in this speech recognizer which are generating the voiceprint and store it as template in the word bank and recognizing the word spoken by comparing it with the template stored in word bank. The voiceprint is create by extracting it’s MEL Frequency Cepstral Coefficient (MFCC) which is the default number of coefficients need to be extracted are 12. To get more accurate result, 20 coefficients will be extracted. For recognizing purpose, Dynamic Time Warping (DTW) method is use. DTW will calculate the distance of two vectors which are the word spoken and the stored voiceprints and it will recognize the word as the same word if the the distance is the lowest or in other words, nearly zero.