Você está na página 1de 7

ACKNOWLEDGEMENT

We convey our special thanks to all who have supported us and helped us in successfully
completing our project.
We are highly indebted to RESHMA VERMA Assistant professor in Department of
Electronics and Communication, for guidance and constant supervision as well as for
providing necessary information regarding the project and also for his support to realize,
implement and improve our project.
We thank our HOD Dr. S. SETHU SELVI, Electronics and Communication Department
having helped us in every way.
We wish to extend our gratitude to respected Principal Dr. N. V. R. NAIDU, MSRIT for
providing the time and facility to complete the project.
We would also like to thank and acknowledge the momentary funding received from Alumni
Association, MSRIT.
We would also like to thank our Institution and faculty members without whom this project
would have been a distance reality. We would also extend our heartfelt thanks to our families
and well-wishers.

MOHAN KRISHNA R
PAVAN KUMAR N
PRASHANTH NATRAJ B
SAURABH KUMAR

Page i

ABSTRACT

Ralph Waldo Emerson: Good thoughts are no better than good dreams, unless they be
executed.
Vince Lombardi: Dictionary is the only place that success comes before work. But in reality
hard work is the price we must pay for success.

We have based our project on these two ideologies. We set out to make a voice
controlled wheelchair. We were successful in creating a prototype of it using HM2007 IC and
interfacing it with microcontroller. We then went a step further and tried simulating the
functionality of the voice recognition IC using GUI (graphical user interface) in MATLAB
and thus making processing real time. On doing so we obtained an accuracy higher than the
IC we used in our project. In the course of this project we have gained hands on experience
with two development environments KEIL and MATLAB. Given more time and guidance we
would want to transfer the code into a DSP processor (thus taking all the processor
parameters in consideration) and then interface it with micro-controller. On the hardware
front we would like to create a ready-to use wheelchair (dealing with the power issues, size
and other system complexities) thus making our system completely indigenous.

Page ii

TABLE OF CONTENTS
Acknowledgement

i.

Abstract

ii.

Table of Contents

iii.

List of Figures

v.

List of Tables

vii.

1.

INTRODUCTION

2.

OVERVIEW OF MICROCONTROLLERR AND ITS PERIPHERALS

3.

4.

2.1. Microcontroller used

2.2. Features of 8051

2.3. 8051 Microcontroller Hardware

2.4. DC Motor Working and its Application in our project

14

2.5. LCD and its programming

19

2.6. 8051 Interrupts

25

HM2007 AND USING KEIL

28

3.1. Brief Description

28

3.2. Speech recognition system using HM2007

29

3.3. HM2007 How is it used?

30

3.4. Steps in executing the C code

33

3.5. A Glance Through the Stages of Development of our Project

34

SPEECH PROCESSING FUNDAMENTALS


4.1. DSP Fundamentals

5.

36
36

SIMULATION OF THE FUNCTIONALITY OF HM2007 USING


MATLAB

44

5.1. Speaker Identification

44

5.2. Feature Extraction

45

Page iii

5.3. Using Vector Quantization

6.

51

CREATING GRAPHICAL USER INTERFACES


6.1. Brief insight on how to create GUI

7.

54
54

IMPLEMENTATION OF SPEAKER IDENTIFICATION IN MATLAB


7.1. Steps for implementation

58
58

8.

RESULT

61

9.

CONCLUSION

63

10.

FUTURE WORK

64

11.

REFERENCES

65

12.

APPENDICES

66

Appendix 1: Datasheet of HM2007

66

Page iv

List of Figures

Fig 2.1: Microcontroller (P89v51RD2) block diagram

Fig 2.2: Pin diagram of 8051

Fig 2.3: Port 0 Diagram

10

Fig 2.4: Port 0 with pull Up Registers

11

Fig 2.5: Port 1 Diagram

12

Fig 2.6: Port 2 Diagram

12

Fig 2.7: Port 3 Diagram

13

Fig 2.8: Forward Movement of the motors

14

Fig 2.9: Reverse Movement of the motors

15

Fig 2.10: The H-Bridge circuit diagram

16

Fig 2.11 the L298 driver board

17

Fig 2.12: Top view of L298 board

18

Fig 2.13: The LCD

20

Fig 2.14: Displaying Kannada Words on LCD

24

Fig 2.15: Activation of INT0 and INT1

25

Fig 2.16: TCON (Timer/counter) Register (Bit-addressable)

26

Fig 3.1: The HM2007 circuit diagram

28

Fig 3.2: The circuit Board for HM2007

29

Fig 3.3: Training of HM2007

34

Fig 3.4: Final assembly and test of programme

35

Fig 4.1: Discrete Fourier Transform

39

Fig 4.2: Diagram of Speech Chain

41

Fig 5.1: Enrollment Phase

44

Fig 5.2: Identification Phase

45

Page v

Fig 5.3: Feature Extraction steps

47

Fig 5.4: Computation of Mel-cepstrum

49

Fig 5.5: Filter Bank in Mel frequency scale

50

Fig 5.6: the vectors generated from training before VQ

51

Fig 5.7: the representative feature vectors resulted after VQ

52

Fig 5.8: Conceptual diagram illustrating vector quantization codebook formation.


One speaker can be discriminated from another based of the location of centroids. 53
Fig 6.1: The GUI layout Editor

55

Fig 7.1: Overview of the Speaker Identification GUI

59

Fig 7.2: Enrollment of Speaker

60

Page vi

List of Tables

Table 2.1: SFR names and equivalent RAM addresses

10

Table 2.2: Alternate Functions of Port 3

13

Table 2.3: The Switch Control

16

Table 2.4: Motor Operation Based On Driver

18

Table 2.5: Pin Description

21

Table 2.6: LCD Command Codes

23

Page vii

Você também pode gostar