Motion and voice interfaces

gaybayberryAI and Robotics

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

139 views





























Migle Maciunaite


Motion and voice interfaces


Helsinki Metropolia University of Applied Sciences

Bachelor of Engineering

Media Engineering

Repor

2011.10.11



Abstract

Author(s)

Title


Number of Pages

Date

Migle Maciunaite

Motion and voice interface


8

pages

11 October

201
1

Degree

Bachelor of Engineering

Degree Progra
m

Media Engineering

Instructor(s)


Jaana Holvikivi,
Doctor of Science

Ulla Paatola, Lecturer
































Keywords

Motion,
voice, interfaces, computers, media
, kinect, wii, apple




Syllabus


1

Introduction

1

2

Motion and voice interfaces

1

2.1

Motion sensors

1

2.1.1

How do motion sensors work?

1

2.1.2

Motion interface products in the marketplace

2

2.1.3

Specifications of Xbox 360 Kinect and Nintendo Wii

3

2.2

Voice

5

2.2.1

How does a voice sensor work?

5

2.2.2

Voice interface products in the marketplace

6

3

Conclusion

7

4

References

8



1


1

Introduction

In the 21
st

century o
ne of the
fastest
growing technological
fi
elds are

motion and voice
sensors which help

us to do basic tasks easier and faster in our daily routine. There are
few companies which could be described as

leaders in today’s industry. Speaking of

voice interface Android and Apple are
the biggest competitors.
Heading side with

each
other in motion

video game console marketplace

there are Microsoft

s Xbox K
inect and
Nintendo Wii.

In the further chapters I am going to
explain

how motion and voice se
n-
sors work and I will compare some of the products that are now in the marketplace.

2

Motion and voice in
terfaces

2.1

Motion sensors



T
here are two kinds of motion detector sensors: Acti
ve and passive. They are sorted
on how they find motion.


2.1.1

How do

motion sensor
s

work
?



For instance, ‘‘ac
tive motion sensors work by sending out bursts of ultrasonic sound
waves
, after which the sensor waits for the energy to be reflected back’’ [1].
These
sensors are mostly used f
or daily

environment such as garage doors, motion detector
lights
, grocery store, floodlights, doorbell

[2].


Passive infrared motion detectors find
and measure
incoming infrared energy” [1]. In simple terms, when the sensors light
are disturbed by a person, the sensor will alert the control box [1]. Those kinds of d
e-
tector
s are used for home, government and

bank
security.



The infrared light is being
disturbed by a person, sending
alarm to the control box [3].

2



Picture

1.
http://static.howstuffworks.com/gif/burglar
-
alarm
-
motion
-
2.jpg

[3].


Now that we know the very basics of how the motion sensors work, we can
move on
to the marketplace of these products.


2.1.2

Motion interface products in the marketplace



In addition to, there are two big names
in the trace of success: Nintendo Wii and M
i-
crosoft’
s X
box
K
inect. They both are using motion interf
ace gaming consoles
which
allow

users to control
a
computer in
the
most natural way


hand movements. They
work a bit differently, with Nintendo Wii you have to use
a
motion

system controller,
which is similar to a TV re
mote. Kinect is unique

and new since you do not need to
use
any controller
, it responses to how you move.
As well as that, you can use voice co
m-
mands with

Kinect.

This product is used not only for games, for example

the

Lithuan
i-
an
company “Etronika” has

created an application which allows

doing

internet banking

operations
using Kinect. The application was first presented at CTIA Enterprise

& A
p-
plications 2011 which took

place in San Diego, also
known

as capital of Silicon Valley,
USA [4].



Products

Xbox Kinect

Nintendo Wii

Console bundle price

$299 (4GB)

$200

Wii motes

-

$40 each

Nun chuck

-

$20 each

Regular game price

$50 each

$49.99

Wii Wheel

-

$15

Wii Zapper

-

$19

Table 1. Comparison of Xb
ox Kinect and Nintendo Wii prices

[5].


Even though the Kinect

gaming console costs more,

it does not require any additional
accessories as Wii does.



3


2.1.3

Specification
s

of
Xbox 360
Kinect and
Nintendo
Wii


Both Wii and Kinect are the products
of seventh generation consoles.

Nevertheless,

that does not mean they have the same features.


Name

Xbox 360 + Kinect

Nintendo Wii

Media

DVD
-
DL

Wii


optical disc

CPU (central processing unit)

3.2

GHz IBM PowerPC tri
-
core
codenamed "Xenon"

729

MHz PowerPC based IBM
"Broadway"

GPU (graphics processing
unit)

500

MHz codenamed "Xenos"
(ATI custom design)

243

MHz ATI "Hollywood"

Memory

512

MB GDDR3 @ 700

MHz
shared between CPU & GPU

10

MB EDRAM GPU frame
buffer memory

24

MB "internal" 1T
-
SRAM
integrated into graphics pac
k-
age

64

MB "external" GDDR3
SDRAM

3

MB GPU frame buffer
memory

User interface

Xbox
360 Dashboard

New Xbox Experience (NXE)

Wii Menu

Input/output

IrDA
-
compliant infrared

for
remote

2 Memory Card slots

3 USB

2.0 ports

1 Ethernet port

Bluetooth 2.0

2 USB 2.0 ports

Four controller and two
memory card ports (Gam
e-
Cube)

1 SD(HC) Card slot

Res
olutions

HDTV
-
capable (480i, 480p,
576i
(50

Hz)
, 576p, 720p,
1080i, 1080p)

Various monitor resolutions
available via VGA and
HDMI/DVI

EDTV
-
capable (480i, 480p,
576i)

Audio

Dolby Digital, WMA Pro,
DTS*, DTS
-
ES*

*
(DVD and HD DVD movies
only)

Dolby Pro Logi
c II surround,
stereo sound and an additio
n-
al Mono speaker is built into
the controller.

Network

100BASE
-
TX Ethernet

Optional 802.11a/b/g/n Wi
-
Fi
adapter (Built in with the Slim
models)

Built
-
in 802.11 b/g Wi
-
fi

Optional Ethernet via USB
adapter

Storage

Included/Optional

detachable
SATA upgradeable 20 GB, 60
GB, 120 GB or 250 GB hard
drive.

Xbox 360 memory cards

USB mass storage (AV co
n-
tent)

512

MB built
-
in flash memory

SD card (up to 32

GB with 4.0
software)

Nintendo GameCube Memory
Cards

The Wii Remote
contains a
16

KiB EEPROM chip from
which a section of 6

kilobytes
can be freely read and written


Table 2.
Xbox 360 and Nintendo Wii

specifications
[6].


4


Furthermore,
Nintendo Wii controller
is wireless and works through Bluetooth
. When
the users
moves
his/her

hand, the motion sensor recognizes it and transmits the signal
to

the

con
soles sensor bar which you have to place near TV. The Wii Remote

collects
the movements of the user, creates a new space in the
boundaries

of the game
and
reflects

them on the

screen [7].
Wii Menu
interface (channels) is navigated using the
pointer
command

with the remote
controller

(pressing and holding A and B buttons).


Picture 2.
The Wii Menu interface with channels explanations [8].


Xbox 360 Kinect is a newer product, w
hich is why they decided to c
hange hand
-
held
controllers to
natural human body commands.
You simply step in

front of the motion
-
sensor, which

tracks 48 points of movement, recognizes your face and hears your
voice.
Kinect has
a
sensor and

a camera that

has

color and depth sensing lenses (Color
VGA motion camera 640 x 480 pixel resolution @30 FPS, Depth Camera 640 x 480
pixel resolution @30 FPS), voice microphone and array supporting single speaker voice

recogni
tion, tilt
motor for sensor adjustment
. [9]



5


Picture 3. Interface of Xbox 360 Kincet [10].


Finally, w
hat makes the

Xbox Kinect special is the way it lets the players get access to
a simple interface that is
proficient

of reading the movements of the body and voice
or
ders [11].

2.2

Voice


The most
natural

way of
using any

device is

simply
by talking with it
.
People talk
faster than they type


this is one of the reasons for voice co
mmanding.


2.2.1

How does a voice sensor work?


First of all, w
hen a command has

been given by voice, the microphone converts

the
analog signal of voice into digital pieces of data which computer

then analyzes. From
that
,
the

computer must take enough information to know the word which was said to
it.
Phonemes are used to form a word for the computer, (i.e. aa


father; ae


cat
; ao


dog.) they make a good data for speech engines to operate with.
“Phonemes are
often extracted by running the waveform through a Fourier Transform. This allows the
waveform to be a
nalyzed in the frequency domain
” [12].




Spectrograph of the "ss" bit of a
ss
ure (this is a phoneme) [13].

Picture 4. http://www.generation5.org/content/2002/images/assure
-
spectrograph.jpp [13].


It is comparatively easy to sync “the amplitudes and frequencies of a template ph
o-
neme with the corre
sponding phoneme in a word” [12].

6


2.2.2


Voice interface products in the marketplace


In 2010 the Android

Company

released a free speech recognition application that a
l-
lowed you to
call,
send texts and email, listen to music, browse the web

just
by spea
k-
ing
into your phone

[14]. This opened a
new era for mobile usage. One year later,

in
2011 Apple released
IPhone

4s with
artificial intelligence called Siri. Unlike with A
n-
droid, you do not need to use 9 prewritten commands, instead you can just speak to
Siri i
n
a
simple everyday language.
“Siri uses the processing power of the dual
-
core A5
chip in iPhone 4S, and it uses 3G and Wi
-
Fi networks to communicate rapidly with A
p-
ple’s data canters” [15]. Both of these products work only with specific mobile devices.
Fe
w days after the big release of Siri an application for every newer smartphone called
Vlingo showed up in the marketplace. It is a virtual assistance that uses voice to text
technology with its "intent engine"

[16] and it can work on most BlackBerry phones
,
Apple's iPhone, iPod Touch, Android phones that run OS 2.0 or higher, Nokia phones
and select Windows Mobile devices.

Name

i
Phone 4s

Android

Vlingo

Unlocked

16GB $649.00

HTC Sensation
4G
$799.99

-

Application

Siri beta

Android vs. 2.3 Gi
n-
gerbread

Vlingo plus max
$19.99 (Free for A
n-
droids)

Table 3. Comparison

of the prices of voice commands

products [17].

Accordingly,

prices vary in different countries
,

although,

the cheapest way to have a
voice interface in your phone is by getting Vlingo application (if your phone supports
the software).


7


Picture 5. Vlingo user interface advertisement [18].

3

Conclusion


In conclusion, v
oice and motion sensors are constantly taking bigger roles in our lives.
When knowing how it works, every person can think of a computer command that
could be done with speech
recognition or hand movements, which

is how they are

b
e-
ing improved.

P
eople

can
now make entertainment even more enjoyable,
business
above accurate
and medicine
outstanding
. As helpful as it may seem,
a lot of topics
always have the down side. When the simplest tasks are being done so fast people
might start losing the sense of time.
Will that make us inefficient workers or
in opp
o-
site
, as we will not be bothered with usual actions our brains will do significant
effort?






8


4

References


1.

Sanders

April. How Does a Motion Sensor Work? [online]. eHow Contributor.

U
RL:
http://www.ehow.com/how
-
does_4596955_motion
-
sensor
-
work.html
.

A
ccessed
17
th

October 2011
.


2.

Kayne R. What is a Motion Detector? [online] Conjecture Corporation.

URL: http://www.wisegeek.com/what
-
is
-
a
-
motion
-
detector.html
.

Accessed 17
th

October 2011
.


3.

How b
urglar alarm works.

[online]


URL: http://home.howstuffworks.com/home
-
improvement/household
-
safety/security/burglar
-
alarm2.htm.

Picture 1 [online].

URL:
http://static.howstuffworks.com/gif/burglar
-
alarm
-
motion
-
2.jpg

Accessed 17
th

October 2011
.


4.

Etronika. ETRONIKA shows off Kinect Banking App. [online]. 14th October
2011.

URL: http://www.etronika.lt/en/news/85/27/ETRONIKA
-
Shows
-
off
-
Kinect
-
Banking
-
App.html.

Accessed 17
th

October 2011.

5.

Table 1

[online].

URL:
http://mashable.com/2010/07/20/kinect
-
p
rice
-
2.

URL:
http://wii.ign.com/launchguide/hardware1.htm
l.

Accessed 17
th

October 2011
.

6.

Table 2

[online].

URL:
http://wii.ign.com/launchguide/hardware1.html
.

URL:
http://www.play.com/Games/Xbox360/4
-
/10296372/Kinect/Product.html
.

Accessed 17
th

October 2011
.


7.

Amy McDonough. Wii Get it Now: Technical Specs, Inside the system and u
n-
der the hood. [online] 11 June 2006.

URL: http://www.1up.com/features/all
-
about
-
wii.

Accessed 17
th

October 2011
.


8.

Picture 2
.

URL:
http://iwantawii.northreviews.com/discover.html.

Accessed 17
th

October 2011


9.

Sriharsh Biswal. Xbox 360 Kinect review, specification and price. [online].

11 February 2011.

URL:
http://www.techibuzz.com/xbox
-
360
-
kinect
-
review/
.

Assessed 17
th

October 2011.


9


10.

Picture 3

[online].

URL:

http://www.xboxpassion.fr/images/xbox/1806/521806
-
kinect
-
interface
-
xbox.jpg.



Accessed 22
nd

October 2011
.


11.


Microsoft xbox Release Kinect vs Nintendo Wii, B
est
Games for Kinect blog

[online]. 2010


Xbox Kinect


Experience the innovative era of video Ga
ming
Units; 2010 August 24
th
.

URL: http://www.zawed.com/kinect
-
interface/.

Accessed 22
nd

October 2011
.


12.


James Matthews. How does speech recognition work? [online]. Generation 5;
October 2002.

URL: http://www.generation5.org/content/2002/howsrworks.asp.

Accessed 22
nd

October 2011.


13.


Picture 4
[online].

URL:
http://www.generation5.org/content/2002/howsrworks.asp


Accessed 22
nd

October 2011.

14.


Steven Mostyn. Android gets impressive Voice Actions speak
-
only interface.
[online]. The Tech Herald; August 2010.

URL: http://www.thetechherald.com/article.php/201032/6022/Android
-
gets
-
impressive
-
Voice
-
Actions
-
speak
-
only
-
interface.

Accessed 22
nd

October
2011.


15.


IPhone 4s: About Siri. [online] Apple inc; October 2011.

URL:
http://support.apple.com/kb/HT4992
.

Accessed 22
nd

October 2011.


16.

Vlingo inc. What is Vlingo? [online]. October 2011.
URL:
http://www.vlingo.com/
.

Acc
essed 22
nd

October 2011.


17.


Table 3
[online].

URL:
http://www.vlingo.com/apps/android
.

URL:
http://store.apple.com/us/browse/home/shop_iphone/family/iphone
.

URL:
http://www.amazon.com/HTC
-
Sensation
-
Z710E
-
Unlocked
-
Smartphone/dp/B005345986/ref=sr_1_1?s=wirel
ess&ie=UTF8&qid=13192912
38&sr=1
-
1
.

Accessed 22
nd

October 2011.



18.

Picture 5

[online].

URL: http://tech.widgetstudy.com/vlingo
-
virtual
-
assistant
-
for
-
android
-
updated
-
with
-
a
-
new
-
user
-
interface
-
and
-
other
-
features/.

Accessed 24
th

October 2011.


Appendix 2

1

(
1
)