Speech Recognition Based Automated Wheelchair Biology Essay

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

This work presents a new idea of how to automate the wheelchair by using finger tracking and speech recognition techniques. The main idea of this inspiration is to process analog finger movement and voice signals and controls the movement of the wheelchair accordingly. The theme is accomplished by implementing both the finger movement tracking system and speech recognition system on the wheelchair to facilitate the paralyzed and handicapped persons to operate the wheelchair conveniently. A unique speech recognition algorithm is used for fast recognition and to make this work real time. The TMS320C6711 DSP Kit is used to process the voice commands and the AVR microcontroller is used to process the dynamic finger movement signals and control the D.C Motors as per instruction. The finger tracking system tracks the movements of a thumb and the first two fingers and the speech recognition system uses the four words as voice commands which are "left", "right", "up" and "stop" to control the motion of the wheelchair. The results at the end show the efficiency of the system.

Keywords:

Wheelchair, dynamic finger movement tracking, speech recognition, movement tracking, gesture recognition, resistive bend sensors, digital signal processing, analog to digital converter (ADC), energy, threshold value, point of maxima, normalization, accumulator, fingerprints generation, fingerprints processing, dictionary, band pass filters, finite impulse response (FIR) filter, Blackman window, coefficient of correlation.

1.1 INTRODUCTION:

Figure 1: Wheelchair

A wheelchair is a mobility device in which the disabled user sits[1]. The People that cannot walk or the people that feel difficulty to walk due to illness, injury, or disability use wheelchairs[1]. People with both walking and sitting disability often need to use a wheelchair[1]. The wheelchair is shown in figure 1. The earliest record of the wheelchair dates back to the 6th century, as an inscription found on a stone slate in China. Later dates relate to Europeans using this technology during the German Renaissance [1].

A mechanical engineer named Harry Jennings, together with his disabled friend and fellow engineer named Herbert Everest, who had broken his back in a mining accident, invented the first lightweight, steel, collapsible wheelchair in 1933[1]. They then started to manufacture the wheelchairs and become the first mass-manufacturers of wheelchairs device[1]. Their "x-brace" design is still common in use, albeit with updated materials and other improvements[1].

This work is aimed specially to encourage those persons who are completely paralyzed or disabled. In this work, the wheelchair is controlled by finger movement tracking, by voice commands (such as Left, Right, Up and Stop) or by joystick. The user can use any of the three systems to instruct the wheelchair to move in the desired direction. Three keys are provided to select either the finger movement tracking system, voice recognition system or joystick.

To track the movement of the fingers, the finger movement tracking technique is used instead of gesture recognition. The difference between the two is given in section 1.2. To recognize the voice commands, the voice recognition system is implemented into the wheelchair. The voice recognition system is implemented by using the TMS320C6711 DSP kit. A detailed description of this unique technique is given in the section from 3.1 to 3.8.

To move the wheelchair, two D.C motors are used along with the gears. The gears provide the necessary torque and improve the load-torque characteristics of the system. The D.C motors are assembled with the rear wheels of the wheelchair. When one D.C motor turns off while the other is running, the torque will be produced and wheelchair will turn towards the side of the stopped motor. When both D.C. motors are running, the wheelchair will move in a straight direction. For switching the D.C motors turn on and off, the relays are used. The AVR microcontroller does not provide the enough current to operate the relays therefore, the transistors are used to amplify the current and operate the relays.

Finally, the results are shown in sections 7.1 and 7.2 that show the efficiency of the system.

1.2 MOVEMENT TRACKING VS. GESTURE RECOGNITION:

Motion tracking is chosen just because of its superiority over gesture tracking as it has no boundaries in order to capture any motion rather than to wait for a special sign being made in gesture tracking.

In gesture recognition, gestures are preprogrammed in the application (i.e., a finite set of very specific body positions) and the user is trained to use these predefined gestures to control the system. The gestures are static, discrete, and restrictive.

As a difference, movement tracking involves constant monitoring and dynamic and continuous tracking of the user who uses a series of movements to continue with the application. In movement tracking, the application is to be trained instead of the user.

Movement tracking emphasizes the natural interaction between the user and application. To achieve natural interaction, the application must learn to adapt to the user's movement rather than the user trying to satisfy the requirements and limitations of the application.

1.3 SYSTEM OVERVIEW:

In this work, a simple wheelchair is automated by using microcontroller, digital signal processor, gearbox, and D.C. motors, controlling circuitry and sensors. The main theme of this work is to control the wheelchair according to the input signals. The input signals can be the voice signals, the signals generating by moving the fingers or the joystick signals. This is illustrated in the figure 2.

The input signal source is selected by pressing the appropriate push buttons. Three push buttons are provided to select the corresponding input signal source but one source at a time. These push buttons are interfaced with the AVR microcontroller through the interfacing circuitry. The digital joystick is also connected with the AVR microcontroller and the user can operate the wheelchair by using digital joystick module.

In this work, the TMS320C6711 DSP kit is used for speech recognition task. It processes the speech signals and generates appropriate analog signals to command the controller so as to control the motion of wheelchair. Four words Left, Right, Up and Stop are used as voice commands.

Special sensors are used to detect the movement of the fingers. These sensors generate analog signals, which are converted by the ADC and translated by the microcontroller to control the motion of wheelchair. Two monitoring sensors are also used to detect the obstacle and uneven surfaces and control the speed of the wheelchair accordingly. LCD module is also used for displaying task. The AVR microcontroller is used for controlling the D.C. motor derive circuitry and LCD, tracking of fingers movement and processing the signals from Digital joystick module.

Figure 2: Block Diagram

Figure 3: Gloves fastened with Resistive

Bend Sensors

Figure 4: Resistive Bend Sensor

2.1 DYNAMIC TRACKING OF FINGERS MOVEMENT:

In this work the first two fingers and a thumb are used to control the wheelchair. The thumb control the speed and forward motion of the wheelchair while the first finger and the second finger control the left and right directions of the wheelchair respectively. For the purpose of tracking the movement of fingers, the resistive bend sensors are used. The bend resistive sensors are fastened at the finger position on the gloves as shown in figure 3.

2.2 RESISTIVE BEND (FLEX) SENSORS:

The resistive bend sensors are the sensors that change their resistance when they are bent. An inflexed sensor has a resistance of about 10,000 Ohms (that is, at 0 degree, its resistance is 10,000 Ohms.) As the sensor is bent its resistance increases to about 35,000 Ohms at 90 degrees as shown in figure 4[2].

Figure 5: Basic Flex Sensor Circuit

As mentioned previously, one thumb and two fingers are used to control the wheelchair so three sensors are used for this purpose. The circuit for measuring the movement of the finger is shown in figure 5.

2.3 FLOW CHART FOR THE FINGER MOVEMENT TRACKING:

The finger movement tracking system starts when the user selects the finger movement signal source as the input source via the push button.

The flow chart is shown in figure 6. The System first scans the thumb sensor input and stores the result in variable A. It then scans the sensor's value for the first finger and stores it in the variable B. It then scans the second finger sensor's value and stores it in the variable C. After scanning all the Resistive Bend Sensors value and storing in the variables, the processing of variables take place. The processing involves the comparing of the values stored in the variable against the values obtained during training. The signals are then sent to the D.C. Motor Drive Circuitry according to the processing results.

It again scans the values of the sensors, processes it and sent the signals to the D.C. Motor Derive Circuitry and so on. The system is in infinite loop and continues until the user selects either the voice command input source or digital joystick input source via the push buttons.

Figure6: Flow Chart for the finger movement tracking system.

3.1 RECOGNITION OF VOICE COMMANDS:

Speech recognition has been actively studied since the 1960s, however recent developments in computer and digital signal processing technology have improved speech recognition capabilities. Speech recognition is very complex problem[3]. It involves many algorithms, which require high computational requirements[3]. Real-time digital signal processing made considerable advances after the introduction of specialized DSP processors[4].

3.2 SYSTEM MODEL:

The model for the speech recognition task is shown in figure 7. The Speech Recognition task is performed by using TMS320C6711 DSP Kit.

3.3 DETECTION OF THE START OF THE WORD:

During training, the DSP Kit calculates the average energy of the noise of the environment and uses it as a threshold point for detecting the start of the word. When the energy of the incoming samples exceed from this threshold value, the system thinks as the word is being spoken and stores the first 3000 samples.

(1)

3.4 DETERMINATION OF THE POINT OF MAXIMA AND SAMPLES NORMALIZATION:

Sometimes we speak louder and sometimes we speak gentle. The energy of the word spoken loudly is much more greater than that of spoken gently. So, there is a need for normalization. One method of normalization that is used in this work is to determine the maximum value in the samples and then divide each sample with this value. That is what called the normalization. After normalization, the samples are multiplied with a certain gain value.

Figure 7: Model For Speech Recognition

3.5 BAND PASS FILTERS:

The FIR filter type is used for Band Pass filter and the Blackman Window is chosen for the FIR filter. The window function[5] is given as:

(2)

The common features of the Blackman Window are described in table 1[5]. The filters are designed in Matlab and the coefficients of these filters are also generated by using Matlab.

Table 1: Features of the Blackman Window

Transition Width (Hz) (Normalized)

Pass Band Ripple (dB)

Main Lobe Relative To Side Lobe (dB)

Stop Band Attenuation

(dB)

5.5/N

0.0017

57

75

3.6 FRAMING:

The samples from each Band Pass filter are grouped into frames. The length of each frame is 200 (that is, there are 200 samples in each frame). The total number of samples is 3000 therefore the number of frames from each filter is 15 (3000/200=15).

3.7 FINGERPRINT GENERATION:

The fingerprint generation is the most important step in the speech recognition task. In this step, the samples are squared and accumulated together for each frame from each filter. It is more easily understood by looking at the following equations:

From Filter 1:

(3)

From Filter 2:

(4)

From Filter 3:

(5)

From Filter 4:

(6)

From Filter 5:

(7)

From Filter 6:

(8)

Figure 7: Model For Speech Recognition

Whereas, N is the number of frames and N=1,2,3…15. So, there are 15 fingerprints for filter1. Similarly, there are 15 fingerprints for filter2, 15 fingerprints for filter3, 15 fingerprints for filter4, 15 fingerprints for filter5 and 15 fingerprints for filter6. So, there are 90 fingerprints(15*6=90) for each word. And as previously mentioned that there are four words(Left, Right, Up and Stop) so, the dictionary contains four words with each having 90 fingerprints.

3.8 FINGERPRINT PROCESSING:

In this step, the fingerprints of the spoken word are compared with fingerprints of the words in the dictionary. Two techniques are combined for achieving the recognition task. First, the system calculates the corresponding difference between the fingerprints of the spoken word and the fingerprints in the dictionary. Lets suppose that [Fd1(1),Fd1(2),Fd1(3),…,Fd1(15)], [Fd2(1),Fd2(2),Fd2(3),…,Fd2(15)], [Fd3(1),Fd3(2),Fd3(3),…,Fd3(15)], [Fd4(1),Fd4(2),Fd4(3),…,Fd4(15)], [Fd5(1),Fd5(2),Fd5(3),…,Fd5(15)] and [Fd6(1),Fd6(2),Fd6(3),…,Fd6(15)] are the fingerprints from filter1, filter2, filter3, filter4, filter5 and filter6 respectively of the word1 in the dictionary and [Fs1(1),Fs1(2),Fs1(3),…,Fs1(15)], [Fs2(1),Fs2(2),Fs2(3),…,Fs2(15)], [Fs3(1),Fs3(2),Fs3(3),…,Fs3(15)], [Fs4(1),Fs4(2),Fs4(3),…,Fs4(15)], [Fs5(1),Fs5(2),Fs5(3),…,Fs5(15)] and [Fs6(1),Fs6(2),Fs6(3),…,Fs6(15)] are the fingerprints from filter1, filter2, filter3, filter4, filter5 and filter6 respectively of the word spoken. Their difference is calculated in the following way:

For Filter 1:

(9)

For Filter 2:

(10)

For Filter 3:

(11)

For Filter 4:

(12)

For Filter 5:

(13)

For Filter 6:

(14)

Where, N=1,2,3…15. In this way the difference between all the four words in the dictionary and the word spoken is computed. The result is four sets of elements with each set consist of 90 elements. Then, the values of all four sets are compared correspondingly with each other and the value with the smallest magnitude is marked. Finally, the marked values are counted and the set with the largest count value is selected. The dictionary word, which is represented by the selected set, is the word that has been selected as the recognized word. For more detail, read the section 7.2.

The second technique used in this work is to calculate the correlation between the spoken word and all the four words in the dictionary. As a result, the four values are generated. All these four values are compared with each other and the value with the largest magnitude is selected. The dictionary word, which is represented by the selected correlation value, is the word that has been selected as the recognized word.

(15)

Where, x is the fingerprints of the dictionary word and y is the fingerprints of the spoken word.

If the two results point to the same dictionary word, the system thinks as it recognizes the word otherwise, the system discards the spoken word, stop the wheelchair and wait for the next voice command.

5.1 JOYSTICKS AND KEYPAD:

The digital joystick and keypad are interfaced to the AVR microcontroller by using the interfacing circuitry. The interfacing circuitry consists of IC "74C922". The IC is interfaced with the AVR microcontroller in the hardware interrupt mode. In this mode, whenever there is any activity on the joystick or any key has been pressed, the IC gives an interrupt to the microcontroller and also gives the binary input for that particular key or joystick movement to the AVR microcontroller.

5.2 MONITORING SENSORS:

For the detection of pit and uneven surfaces, the ultrasonic sensor is used. Also, the ultrasonic transmitter and receivers are used to monitor the distance from the wheelchair to the obstacle if there is any obstacle in the range of the ultrasonic sensors. One property of ultrasonic sensor is that it is not light sensitive. The other property is that the ultrasonic sensors produce mechanical vibrations therefore the ultrasonic sound waves can reflect from the glass surface.

6.1 SYSTEM TRAINING:

Two push buttons for the training of "Finger movement tracking system" are interfaced to the microcontroller. The first key is called the training key and the other key is known as the signal acceptance key. When the training key is pressed, the system is entered into the training mode. In this mode, the system first asks from the user to enter the first angle for the "Left" command by displaying a message on the LCD. When the user, after bending the finger(s), presses the signal acceptance key, the system captures the first angle for the "left" command and stores it. The user can bend the finger(s) at any angle between 0 and 90 degrees. Then system asks for the second angle for the same "left" command from the user. When the user again presses the acceptance key after bending the finger(s), the system captures the second angle for the "left" command and stores it. The system then asks from the user to enter the two angles for the "Right" command, two angles for the "forward" command and two angles for the "stop" command in the same way as in the case of "left" command one by one. In this way, the system is trained for "Left", "Right", "Forward" and "Stop" signals. The reason for the two inputs for every command is to create a threshold window. If the angle appears in between this window, the appropriate command is generated for the wheelchair. If the windows of any two commands overlap, the system displays a message of incorrect input.

There are three useful buttons that are provided in the TMS320C6711 DSP kit. For the training of "Speech recognition system", the first two buttons are used. When the first button is pressed, the system enters into the training mode. In this mode, the system first calculates the average energy of the noise in the environment to set the threshold level and asks from the user to speak the "Left" word for entering its fingerprints into the dictionary by displaying the message on the LCD. When the user speaks the word "Left", its fingerprints are generated and stored into the dictionary and the system asks to speak the next word and so on. In this way, the fingerprints of the words "Left", "Right", "Up" and "Stop" are stored into the dictionary.

7.1 RESULTS OF THE FINGER MOVEMENT TRACKING SYSTEM:

The table 2 represents the angles obtained during system training. These are the angles between which the wheelchair operates.

Table 2: Testing Results

FORWARD

LEFT

RIGHT

STOP

Thumb's Angle

40°-70°

Angle other than

40°-70°

Angle other than

40°-70°

The Wheelchair will be stopped other wise.

1stFinger's Angle

Angle other than

30°-55°

30°-55°

Angle other than

30°-55°

2ndFinger's Angle

Angle other than

35°-60°

Angle other than

35°-60°

35°-60°

7.2 FINGERPRINTS PROCESSING RESULTS AND DISCUSSION:

The fingerprints of the four different words are shown in figure 8. These graphs are generated by using matlab. The table 5 shows the difference between the spoken word "Left" and the words in the dictionary. The table 6 shows the difference between fingerprints of the spoken word "Right" and the fingerprints of the words in the dictionary. The table 7 shows the difference between fingerprints of the spoken word "Up" and the fingerprints of the words in the dictionary. The table 8 shows the difference between the fingerprints of the spoken word "Stop" and the fingerprints of the words in the dictionary. As previously mentioned, the system first compares all the values in a row and marks the smallest value as shown in the table by highlighted values. Finally, count the marked values along each column and the column with the largest value is selected as shown in table 3. The dictionary word, which is represented by the selected column, is the word that might be been spoken. Also, the correlation is calculated between the fingerprints of the spoken word and the fingerprints of all the four words in the dictionary. As a result, the four values are generated. All these four values are compared with each other and the largest value is selected. The dictionary word, which is represented by the selected correlation value, is the word that might be spoken. The results are shown in table 4.

Table 3: Total number of highlighted values in each column in

each table

Total Number of marked values in the column "LEFT"

Total Number of marked values in the column "RIGHT"

Total Number of marked values in the column "UP"

Total Number of marked values in the column "STOP"

FOR TABLE 4

43

10

10

27

FOR TABLE 5

20

46

11

13

FOR TABLE 6

17

10

46

17

FOR TABLE 7

7

11

14

58

Table 4: Correlation result between the spoken words and the dictionary words.

SPOKEN WORDS

LEFT

RIGHT

UP

STOP

DICTIONARY WORDS

LEFT

9.598631e-001

1.584156e-001

2.832750e-001

5.539552e-002

RIGHT

6.775456e-001

9.672273e-001

1.119704e-001

8.441853e-001

UP

2.772080e-001

4.847244e-002

9.060322e-001

9.317274e-001

STOP

5.358149e-001

1.773697e-001

8.197775e-001

9.776034e-001

Figure 8: Fingerprints of four different words

Table 5: Difference between the fingerprints of the spoken

word "Left" and the words in the dictionary.

LEFT

RIGHT

UP

STOP

4.31E-15

1.02E-14

2.8E-15

5.38E-16

5.61E-10

1.09E-09

3.48E-10

1.12E-11

1.61E-06

3E-06

1.12E-06

2.72E-07

0.00044

0.000843

0.000345

0.000234

0.700116

2.28947

0.77494

0.141289

3.338605

35.90434

47.50365

66.54823

2.554634

98.23781

130.8999

97.93874

55.70084

52.26604

91.49724

50.74404

22.49628

39.71779

46.42148

40.70846

71.89665

108.1798

166.5375

116.8931

9.257445

104.8081

168.5496

138.6852

34.06587

9.379772

96.17156

83.91616

4.518435

76.72673

5.459473

0.400824

0.087853

35.6757

0.037299

0.420068

0.038408

11.68717

0.029307

0.002655

1.58E-20

4.49E-20

3.11E-20

7.63E-21

1.11E-17

3.54E-17

5.89E-17

2.77E-17

6.77E-10

1.1E-08

1.36E-08

1.99E-11

3.65E-07

0.000161

0.000379

0.000116

0.030593

0.476143

0.480362

0.174129

0.05912

1.135466

21.9065

12.13776

0.000498

5.880709

30.76166

17.97669

2.090408

9.46717

30.44977

25.33527

2.730677

2.871259

5.852592

21.17592

4.890821

5.29243

9.23342

12.19862

3.99427

9.234598

10.59553

2.290238

3.583455

5.391103

5.910371

1.411196

0.294125

0.261543

0.364755

2.064769

0.006261

0.006571

0.007252

0.431321

0.026747

0.021225

0.016756

0.167252

1.9E-20

3.71E-20

1.68E-20

4.92E-22

1.39E-17

2.47E-17

2.01E-17

6.98E-18

1.76E-10

1.81E-09

5.43E-09

2.29E-10

2.24E-06

3.39E-05

0.000152

1.98E-06

0.005195

0.036722

0.284848

0.005608

0.021314

0.559154

11.84622

1.761157

0.031127

2.786748

12.8614

0.864892

0.272342

2.576011

3.719004

0.27419

0.390814

1.416993

0.784509

0.877976

0.77076

0.460554

1.242732

1.223078

0.054329

1.224042

1.416664

1.38865

0.730466

0.941947

0.985411

0.978062

0.058162

0.046513

0.072179

0.06817

0.02196

0.001633

4.3E-05

0.005407

0.001416

0.019382

0.024237

0.026947

5.11E-21

1.11E-20

3.09E-21

1.03E-22

3.8E-18

7.35E-18

3.25E-18

1.74E-18

2.91E-10

2.19E-09

2.29E-10

9.24E-12

2.63E-06

2.51E-05

1.37E-06

2.69E-06

0.011609

0.100273

0.012662

0.002162

0.005135

0.175292

0.06002

0.046324

0.011497

0.059122

0.075095

0.002008

0.06691

0.140801

0.11192

0.180389

0.270155

0.435585

0.458007

0.465353

0.070635

0.712687

0.737655

0.734284

0.089036

1.153877

1.218951

1.217215

0.441127

0.487822

0.561791

0.560851

0.006077

0.003289

0.020016

0.019814

0.004865

0.001945

0.006395

0.006988

0.000831

0.004326

0.00533

0.005641

3.44E-22

1.62E-21

1.25E-22

5.61E-23

2.3E-19

1.15E-18

1.68E-19

1.74E-19

2.32E-11

7.98E-10

1.97E-10

2.72E-11

6.03E-07

7.99E-06

2.88E-06

5.66E-07

0.001208

0.04679

0.006522

0.000118

0.002403

0.063871

0.077474

0.039023

0.006836

0.033794

0.056919

0.00486

0.007376

0.062986

0.011103

0.007854

0.036706

0.034923

0.008978

0.008433

0.074034

0.006952

0.040379

0.036636

0.074798

0.149956

0.193435

0.189639

0.111774

0.111937

0.193939

0.191409

0.007039

0.047356

0.004678

0.003968

0.000863

0.004158

0.00493

0.005888

0.000395

0.000875

0.004033

0.00401

2.94E-19

5.08E-18

4.77E-19

8.31E-19

1.57E-17

5.1E-16

7.49E-18

3.87E-17

1.86E-11

5.14E-10

3.27E-11

4.51E-11

3.67E-08

4.65E-06

1.52E-07

1.48E-06

0.000584

0.021141

8.52E-05

0.002125

0.000772

0.033486

0.007454

0.016427

0.001172

0.029542

0.002997

0.000373

0.009153

0.000125

0.019073

0.020256

0.047357

9.86E-05

0.013126

0.013711

0.072265

0.004275

0.005519

0.005946

0.036318

0.034704

0.012001

0.012796

0.010686

0.079177

0.006181

0.006203

0.003926

0.122009

0.00092

0.001472

0.000869

0.009003

0.00421

0.004274

4.11E-05

0.001158

0.003166

0.003083

Table 6: Difference between the fingerprints of the spoken

word "Right" and the words in the dictionary.

LEFT

RIGHT

UP

STOP

5.41E-15

3.78E-15

2.36E-15

6.71E-15

1.26E-09

4.57E-10

2.48E-10

8.73E-10

5.44E-06

1.3E-06

3.9E-07

2.62E-06

0.002516

0.000324

0.000192

0.000769

0.275505

0.490503

0.537435

1.127742

49.28251

7.925268

30.27387

56.19407

100.3115

14.56143

72.76977

128.0931

120.6016

19.68241

81.88865

94.40314

9.609918

38.43834

72.00518

39.37976

148.9497

12.1531

155.5255

141.423

171.0216

48.87206

171.0862

150.2696

156.7957

39.46628

156.8011

153.7325

111.5571

25.49659

111.5477

110.8845

38.80267

11.94835

38.79379

38.77955

8.270336

2.536724

8.017183

8.274508

4.92E-20

1.21E-20

1.67E-20

2.54E-20

6.78E-17

7.38E-18

6.09E-18

1.77E-17

8.47E-09

1.7E-09

1.99E-08

7.99E-10

0.000136

1.78E-06

0.000142

2.41E-05

0.22667

0.130027

0.257414

0.050713

4.148318

0.055477

13.59302

7.649203

5.872783

0.247842

18.59614

26.24078

8.012206

1.255107

21.6971

42.27143

2.606584

0.562851

5.790844

9.997974

17.80617

11.3179

18.21498

13.95895

11.58804

7.356134

11.59594

3.31224

3.430674

1.817018

3.432266

2.163896

0.64606

0.306688

0.648212

0.480262

0.090233

0.066572

0.097151

0.086972

0.005238

0.007408

0.041059

0.00727

4.02E-20

1.52E-20

8.21E-21

2.93E-20

6.11E-17

1.03E-17

4.75E-19

2.16E-17

2.48E-10

7.69E-10

9.65E-10

1.98E-10

3.37E-06

9.96E-06

2.14E-05

6.51E-06

0.011227

0.028876

0.015299

0.021955

0.560534

0.109091

2.859315

1.62512

0.707884

0.093202

2.355506

0.656665

0.653758

0.271745

3.044724

0.016125

1.817253

1.475675

2.052637

2.117212

4.254935

1.299772

4.296252

4.289268

3.671693

3.079114

3.683852

3.671314

0.902091

0.738091

0.905939

0.909622

0.129962

0.102132

0.127764

0.133419

0.007111

0.010786

0.018713

0.021183

0.018306

0.004949

0.015225

0.010995

1.11E-20

4.04E-21

1.79E-21

8.07E-21

1.68E-17

2.82E-18

4.12E-19

6.09E-18

1.5E-10

3.89E-10

1.73E-09

1.76E-12

1.87E-06

4.34E-06

1.96E-05

4.95E-07

0.006485

0.021566

0.079157

0.003322

0.419907

0.048818

0.186914

0.062126

0.323897

0.028692

0.041671

0.033665

0.553248

0.028613

0.043049

0.022125

0.300221

0.063296

0.101582

0.087562

0.097128

0.060921

0.152214

0.151097

0.399891

0.354266

0.419115

0.415292

1.515249

1.390927

1.51777

1.519129

1.750389

1.65329

1.754618

1.755272

0.172727

0.177086

0.177768

0.179076

0.015979

0.021379

0.019979

0.0239

9.38E-22

1.44E-22

3.29E-22

3.9E-22

1.31E-18

2.25E-19

3.91E-19

2.45E-19

2.27E-10

3.57E-10

3.11E-10

1.29E-10

2.44E-06

4.17E-06

3.78E-06

4E-06

0.00486

0.020988

0.013918

0.001643

0.03463

0.031865

0.021479

0.171987

0.072805

0.019684

0.029181

0.038243

0.160706

0.048554

0.054974

0.050585

0.036191

0.083425

0.063597

0.056583

0.046008

0.138503

0.066185

0.061592

0.175036

0.038508

0.180292

0.173193

0.385654

0.268145

0.388255

0.385888

1.236967

1.130679

1.24086

1.24025

0.411233

0.408609

0.413362

0.413374

0.0778

0.081913

0.080534

0.082861

2.33E-19

1.66E-18

2.97E-18

9.61E-19

3.69E-17

1.74E-16

3.29E-16

1.4E-16

1.02E-11

2.85E-10

2.32E-10

1.15E-10

1.01E-06

2.83E-06

2.46E-06

6.06E-07

0.004162

0.014222

0.013414

0.005359

0.014304

0.01247

0.013152

0.161099

0.015782

0.050652

0.008366

0.000861

0.019421

0.041459

0.03069

0.03175

0.016504

0.002721

0.029717

0.0296

0.000397

0.037127

0.006562

0.006991

0.011253

0.150373

0.01476

0.014822

0.098509

0.159728

0.100737

0.101044

0.698085

0.509249

0.699181

0.699033

0.238288

0.2074

0.240517

0.240937

0.028789

0.030167

0.028779

0.030833

Table 7: Difference between the fingerprints of the spoken

word "Up" and the words in the dictionary.

LEFT

RIGHT

UP

STOP

6.11E-15

4.57E-15

6.81E-16

4.1E-15

7.43E-10

3.99E-10

5.81E-10

7.87E-10

1.76E-06

5.98E-07

2.79E-06

2.68E-06

0.001054

0.001316

0.001828

0.001887

0.564217

0.374867

0.242046

1.881921

51.87263

52.54753

19.51014

6.956851

104.6847

125.3258

18.42753

40.15437

99.25428

124.1021

81.50662

23.96666

115.8914

97.57564

27.50049

92.73677

118.651

117.39

0.104637

82.19165

75.74452

103.9056

0.026183

25.73891

4.279102

94.95513

0.005056

0.914463

0.12207

57.84628

0.004343

0.028629

0.020742

18.16188

0.013315

0.064477

0.04749

5.109108

0.008766

0.015855

5.62E-22

3.13E-21

9.49E-20

4.3E-20

7.83E-18

1.04E-17

4.6E-17

3.72E-17

2.38E-08

1.83E-08

6.89E-08

2.59E-08

0.000238

0.000174

0.000876

0.000247

0.59736

0.366097

1.62101

0.661962

57.9916

57.60573

6.631295

48.01595

102.777

102.8077

32.72216

74.09062

73.81701

75.38645

50.38311

34.11423

6.13151

7.962048

6.030227

34.30446

6.043575

9.414634

0.064097

25.31616

1.892205

3.434507

0.006915

5.157491

0.102702

0.823882

0.006772

0.04796

0.022373

0.114081

0.000534

0.005914

0.016962

0.025639

0.00122

0.004909

0.02283

0.003418

0.000651

0.001412

2.47E-20

1.33E-20

4.37E-21

2.74E-20

7.91E-18

2.7E-18

9.31E-18

2.95E-17

3.02E-10

1.37E-09

1.98E-08

3.03E-10

6.47E-05

4.64E-05

0.000658

6.72E-05

0.021307

0.067581

0.526835

0.000969

7.894594

7.752463

1.204448

5.509716

11.14613

10.91692

5.501759

9.720994

6.217136

6.399241

6.607147

6.525094

2.768052

4.940542

0.072734

0.009058

1.388632

4.340324

0.005185

0.014454

0.445464

1.307845

0.004136

0.003628

0.02587

0.198081

0.001391

0.00028

0.014998

0.026114

0.000366

0.000152

0.016296

0.001524

0.002703

0.000345

0.023199

0.007342

0.000428

0.000357

7.11E-21

4.11E-21

5.02E-21

6.64E-21

3.97E-18

1E-18

7.93E-18

5.64E-18

3.31E-12

5.64E-10

1.54E-10

4.82E-10

2.59E-06

7.8E-06

5.94E-06

1.14E-05

0.002856

0.03582

0.009793

0.019769

0.144694

0.078039

0.048984

0.096903

0.087913

0.030518

2.8E-05

0.033635

0.330972

0.007189

0.089009

0.057987

2.804238

0.318446

0.003957

0.013009

2.022192

0.162212

0.001013

0.005844

0.289792

0.348651

0.000749

0.001879

0.012428

0.226681

0.000952

0.00189

0.007382

0.087702

8.49E-05

0.001866

0.00956

0.003465

0.001292

0.000966

0.015104

0.002176

0.000479

0.001208

5.56E-22

7.65E-22

5.49E-22

1.78E-21

2.57E-19

3.8E-19

8.35E-19

1.42E-18

2.71E-10

7.51E-10

4.09E-10

9.99E-10

7.75E-07

3.17E-06

1.52E-06

8.48E-06

0.016986

0.043927

0.021653

0.046879

0.012423

0.032205

0.136209

0.129499

0.127568

0.123403

0.03827

0.119151

0.053162

0.076961

0.099039

0.104141

0.259961

0.062112

0.005147

0.008804

0.124275

0.067719

0.000483

0.009449

0.043817

0.096627

6.37E-05

0.004044

0.023559

0.047517

0.000485

0.00095

0.005673

0.021332

1.01E-05

0.000742

0.004152

0.006621

3.84E-05

0.000166

0.005609

0.002055

0.000112

0.000863

4.3E-20

1.83E-18

1.48E-18

1.71E-17

2.5E-17

1.98E-16

1.28E-16

1.67E-15

1.58E-10

6.46E-10

2.07E-11

1.21E-09

1.1E-06

8.96E-06

8.34E-07

1.33E-05

0.005052

0.025435

0.001912

0.073554

0.003272

0.01056

0.007886

0.259531

0.003601

0.010559

0.000112

0.00444

0.016021

0.026998

0.000993

0.000277

0.084366

0.022828

0.000207

3.48E-05

0.028138

0.018988

0.000259

0.000153

0.00654

0.018769

6.38E-05

0.000116

0.001556

0.072642

0.000375

4.44E-05

0.003047

0.053458

0.000223

0.000213

0.004219

0.008823

0.000176

0.000252

0.002408

0.000866

0.000419

0.000151

Table 8: Difference between the fingerprints of the spoken

word "Stop" and the words in the dictionary.

LEFT

RIGHT

UP

STOP

4.16E-16

1.35E-14

4.14E-15

2.62E-16

1.07E-10

2.22E-09

8.19E-10

3.35E-11

7.37E-07

7.24E-06

3.09E-06

2.24E-07

0.000548

0.001846

0.001214

0.000351

0.013512

2.46923

1.124652

0.214962

78.28374

11.80903

31.58129

36.72353

161.7394

34.91665

63.93303

21.31872

163.0723

23.97921

29.57227

18.02676

114.7321

41.86225

77.16904

12.32519

7.834907

101.4614

25.86369

8.853068

19.37549

34.60262

6.996401

13.20255

155.481

5.220456

2.301019

1.376286

158.4518

1.470039

0.518158

0.477769

129.7402

0.554198

0.036695

0.024557

57.07805

0.229488

4.395281

0.009298

2.56E-21

6.32E-20

4.91E-20

1.86E-21

1.29E-17

5.25E-17

3.96E-17

9.29E-19

7.85E-09

1.48E-09

9.98E-09

2.7E-10

5.85E-05

8.65E-06

0.00028

8.5E-06

0.350536

0.042279

0.149447

0.035934

12.6405

4.215925

1.170786

12.21764

73.57624

40.40583

41.1066

11.2113

102.9598

72.88597

58.5796

34.57204

76.25016

64.76683

68.65399

17.68682

12.67227

10.39713

12.81701

6.340333

1.453935

1.627932

1.877288

5.953715

9.098802

1.477901

1.487458

1.334379

11.23217

0.168205

0.172937

0.171037

8.316571

0.04245

0.042046

0.043756

2.715226

0.004286

0.186416

0.007631

1.74E-21

7.47E-20

2.97E-20

1.14E-21

1.08E-17

6.48E-17

2.38E-17

4.9E-18

8.51E-09

2.13E-10

5.83E-10

2.76E-10

9.39E-05

4.82E-06

2.02E-05

1.98E-05

0.27412

0.000703

0.018531

0.000328

2.404509

0.116061

0.385767

0.922001

0.822136

4.349699

1.975474

0.510708

0.396381

6.517968

1.664834

0.138077

0.059119

3.843689

0.119816

0.005803

0.100295

0.275346

0.005612

0.005899

0.38375

0.019276

0.005602

0.005901

3.856632

0.012765

0.005705

0.000971

5.379351

0.007294

0.005975

0.001154

2.425594

0.010926

0.001556

0.001803

0.342863

0.007915

0.04913

0.001696

3.01E-22

2.05E-20

7.5E-21

4.51E-22

3.39E-18

1.82E-17

6.48E-18

6.76E-19

1.16E-09

8.23E-11

3.01E-10

2.14E-10

1.25E-05

2.95E-07

2.99E-06

9.58E-07

0.071156

0.019927

0.020914

0.003677

0.005533

0.1791

0.012347

0.003779

0.091352

0.65914

0.057005

0.034523

0.066636

1.601483

0.06263

0.031272

0.017359

3.153052

0.002318

0.002029

0.013197

1.812856

0.003346

0.002879

0.100612

0.121927

7.28E-05

0.001866

5.856168

0.013007

0.001191

9.44E-05

6.731162

0.002227

0.000996

0.000481

2.730434

0.002157

0.001277

5.49E-05

0.294162

0.002136

0.005707

0.00022

8.71E-22

2.26E-21

8.16E-22

9.25E-23

5.13E-19

2E-18

6.5E-19

1.41E-19

7.96E-10

5.15E-10

1.74E-10

1.06E-10

8.99E-06

7.79E-06

1.64E-06

1.56E-06

0.046371

0.027778

0.009481

0.004154

0.037768

0.007255

0.03923

0.017795

0.093177

0.040104

0.042296

0.050785

0.030587

0.105712

0.068484

0.016582

0.023386

0.45361

0.01415

0.001784

0.00042

0.380064

0.009636

0.006814

0.021087

0.347957

0.00232

0.003295

0.586293

0.015342

0.000217

0.000279

0.96851

0.002198

0.000268

0.000581

0.492995

0.001363

0.000343

0.000403

0.100652

0.000789

0.007786

2.78E-05

5.13E-18

4.05E-18

1.58E-18

1.06E-18

4.33E-16

3.51E-16

1.08E-16

2.93E-17

6.22E-10

2.9E-10

1.87E-10

1.46E-10

5.58E-06

1.42E-06

6.73E-07

3.06E-07

0.026125

0.016799

0.006514

0.003115

0.006543

0.040688

0.010356

0.039689

0.012552

0.002025

0.009948

0.007754

0.004956

0.015063

0.002498

0.003478

0.000253

0.037783

0.000466

0.00043

0.001071

0.1197

0.00013

0.000165

0.006229

0.081435

0.000107

6.24E-05

0.344809

0.004916

6.29E-05

0.000109

0.393801

0.001415

0.000219

0.000148

0.122832

0.001251

0.000246

0.000168

0.005901

0.000866

0.003118

9.35E-05

7.4 CONCLUSION:

Recent improvements in the technology are making lives easier for everybody. This work is to help the paralyzed persons by implementing finger movement tracking and speech recognition system onto the wheelchair to make it an intelligent device. The finger movement tracking system and the speech recognition task has successfully implemented by resistive bend sensors and AVR microcontroller and using TMS30C6711 DSP kit respectively. Furthermore, AVR microcontroller is used to control the motion of the wheelchair.

Writing Services

Essay Writing
Service

Find out how the very best essay writing service can help you accomplish more and achieve higher marks today.

Assignment Writing Service

From complicated assignments to tricky tasks, our experts can tackle virtually any question thrown at them.

Dissertation Writing Service

A dissertation (also known as a thesis or research project) is probably the most important piece of work for any student! From full dissertations to individual chapters, we’re on hand to support you.

Coursework Writing Service

Our expert qualified writers can help you get your coursework right first time, every time.

Dissertation Proposal Service

The first step to completing a dissertation is to create a proposal that talks about what you wish to do. Our experts can design suitable methodologies - perfect to help you get started with a dissertation.

Report Writing
Service

Reports for any audience. Perfectly structured, professionally written, and tailored to suit your exact requirements.

Essay Skeleton Answer Service

If you’re just looking for some help to get started on an essay, our outline service provides you with a perfect essay plan.

Marking & Proofreading Service

Not sure if your work is hitting the mark? Struggling to get feedback from your lecturer? Our premium marking service was created just for you - get the feedback you deserve now.

Exam Revision
Service

Exams can be one of the most stressful experiences you’ll ever have! Revision is key, and we’re here to help. With custom created revision notes and exam answers, you’ll never feel underprepared again.