Seminar Report 2014 Silent Sound Technology

Seminar Report 2014
Silent Sound Technology
CHAPTER 1
INTRODUCTION
The word Cell Phone has become greatest buzz word in Cellular Communication
industry. Silent Sound Technology is an amazing solution for those who had lost their voice but
wish to speak over phone. It enables speech communication when an audible acoustic signal is
unavailable. It is a technology for mobile phones and is definitely going to be a good solution for
those feeling annoyed when other speak loud over phone. This technology transmits information
without using vocal tract and it reduces noise pollution to a great extent. This is very helpful to
the people who are vocally challenged or have been rendered mute due to accidents. Silent
Sound Technology aims to notice every movement of the lips and transform them into sounds.
People suffering from disorders of vocal chords cannot answer any phone calls, so people who
cannot speak can simply give the lip movements according to the words they required to speak
and the silent lip movements are converted into a computer generated sound and transmit
through the mobile phones. It is widely agreed that cell phones can be an annoyance in meetings
or quiet areas and in many public places today their use is banned. Quite often the cell phone
user too, is uncomfortable having the content of his or her conversation become public. At the
same time, the ability to field an urgent or important call at any location could in many instances
be a very useful service. This technology, if non-invasive and small enough to be incorporated
into a telephone handset, would resolve all the issues by allowing user to communicate silently
without disturbing others. Rather than making any sounds, the handset would decipher the
movements made by ones mouth by measuring muscle activity, then convert this into speech
then the person on the other end of the call can hear. The technology opens up a host of
applications, from helping people who have lost their voice due to illness or accident to telling a
trusted friend your PIN number over the phone without anyone eavesdropping, assuming no lipreaders are around. It can be used in Military for communicating secret/confidential matters to
others Also, given the numbers of cell phones in use today, the market for this technology could
potentially become very important if such a concept gained public acceptance. Hence Silent
Sound Technology has a great future.
Dept. of ECE, T.K.M Institute Of Technology, Kollam
Page 1
Seminar Report 2014
1.1 INVENTION
The idea of interpreting silent sound electronically or with a computer has been
around for a long time and was popularized in 1968 Stanley Kubrick science fiction film-2001
A Space Odyssey where a HAL 9000 computer was able to lip-read the conversations of
astronauts who were plotting its destruction. The first true Silent Speech Interface system
which deployed 3 electromyographic sensors mounted on speaker's face and interpreted the
speech with an accuracy of 71%, originated in Japan.
This device is presently being developed by Karlsruhe Institute of Technology (KIT),
Germany by using electromyography. In the year 2010 at CEBIT, one of the largest trade fair
Silent Sound Technology was developed. The device, developed by the Karlsruhe Institute of
Technology (KIT), uses electromyography, monitoring tiny muscular movements that occur
when we speak and converting them into electrical pulses that can then be turned into speech,
without a sound uttered.
1.2 HOW WE SPEAK..?

When we generally speak aloud air passes through the larynx or vocal chords and the
tongue. Words are produced by using articulator muscle in the mouth and jaw region.
Page 2
Seminar Report 2014
Now imagine a technology that could allow everyone to make phone calls and talk
without making a sound. Recently it is proved that articulator muscle become active irrespective
of whether air passes through them or not. The weak signals are sent from the brain to the speech
muscle. These signals are collectively known as the electromyograms. That is, this technology
works only with the movement of facial muscles when we try to speak.
Page 3
Seminar Report 2014
CHAPTER 2
CONCEPT OF SILENT SOUND TECHNOLOGY

You are in a movie theater or noisy restaurant or a bus etc where there is lot of noise
around is big issue while talking on a mobile phone. But in the future this problem is eliminated
with silent sounds, a new technology that detect very lip movement and internally converts the
electrical pulses into sound signals and sends them neglecting all other surrounding noise.
Common people talking at same place without disturbance

Silent Sound Technology aims to notice every movement of the lips and transform them
into sounds, which could help people who lose voices to speak, and allow people to make silent
calls without bothering others. So, basically, it reads your lips. We currently use electrodes
which are glued to the skin.
Even by saying words without producing sound, weak electric currents are sent from your
brain to the speech muscle. These electrical signals are known as electromyograms. These
electric signals play a major role in interpreting silent sounds from silent lip movements as these
electromyographic signals are analyzed and converted to respective voice signals. The awesome
Page 4
Seminar Report 2014
feature added to this technology is that it is an instant polyglot. That is, movements can be
immediately transferred into the language of users choice. This translation works for languages
like English, French and German.
Page 5
Seminar Report 2014
CHAPTER 3
ELECTROMYOGRAPHY
Electromyograpy (EMG) refers to the collective electric signal from muscles, which is
controlled by the nervous system and produced during muscle contraction. The signal represents
the anatomical and physiological properties of muscles; in fact an EMG signal is the electrical
activity of a muscles motor units. It is a technique used in Silent Sound Technology that
monitors tiny muscular movements that occur when we speak and converting them into electrical
pulses that can then be turned into speech without a sound utter. EMG is performed using
instrument called an Electromyograph, to produce a record called an Electromyogram. An
Electromyograph detects the electrical potential generated by muscle cells when these cells are
electrically or neurologically activated. The signals can be analyzed to detect medical
abnormalities, activation level, or to analyze the biomechanics of human or animal. Fig 1 shows
the basic mechanism of Electromyography where the muscle activity is analyzed by the
Electromyograph to generate an Electromyogram movement.
Electromyography
Page 6
Seminar Report 2014
3.1 ELECTRICAL CHARACTERISTICS

The electrical source is the muscle membrane potential of about -90 mV. Measured EMG
potentials range between less than 50 V and up to 20 to 30 mV, depending on the muscle under
observation. Fig 2 shows EMG sensors connected on the face of a speaker. Typical repetition
rate of muscle motor unit firing is about 720 Hz, depending on the size of the muscle. In
interfacing we use four different kinds of transducers
Vibration sensors
Pressure sensor
Electromagnetic sensor
Motion sensor
Electromyographic sensors attached to face

3.2 TYPES OF EMG
There are two kinds of EMG in widespread use: surface EMG (sEMG) and intramuscular
(needle and fine-wire) EMG. To perform intramuscular EMG, a needle electrode or a needle
containing two fine-wire electrodes is inserted through the skin into the muscle tissue. A trained
professional such as a neurologist, physiatrist, chiropractor or physical therapist observes the
Page 7
Seminar Report 2014
electrical activity while inserting the electrode. Certain places limit the performance of needle
EMG by non-physicians. A recent case ruling in the state of New Jersey declared that it cannot
be delegated to a physician's assistant. The insertional activity provides valuable information
about the state of the muscle and its innervating nerve. Normal muscles at rest make certain,
normal electrical signals when the needle is inserted into them. Then the electrical activity when
the muscle is at rest is studied. Abnormal spontaneous activity might indicate some nerve and/or
muscle damage. Then the patient is asked to contract the muscle smoothly. The shape, size, and
frequency of the resulting electrical signals are judged. Then the electrode is retracted a few
millimeters, and again the activity is analyzed until at least 1020 motor units have been
collected. Each electrode track gives only a very local picture of the activity of the whole muscle.
Because skeletal muscles differ in the inner structure, the electrode has to be placed at various
locations to obtain an accurate study. Intramuscular EMG may be considered too invasive or
unnecessary in some cases. Instead, a surface electrode may be used to monitor the general
picture of muscle activation, as opposed to the activity of only a few fibers as observed using an
intramuscular EMG.
Representation of the preferred electrode location

The preferred electrode location is between the motor point (or innervation zone) and the
tendinous insertion, with the detection surfaces arranged so that they intersect as many muscle
fibers as possible.
These days, surface-detected signals are preferably used to obtain information about the
time or intensity of superficial muscle activation. Surface Electromyography (sEMG) is the
process of recording electrical muscle activity captured by surface (i.e.
non-implanted)
electrodes. When a muscle fiber is activated by the central nervous system, small electrical
currents in the form of ion flows are generated. These electrical currents move through the body
Page 8
Seminar Report 2014
tissue, whose resistance creates potential differences which can be measured between different
regions on the body surface, for example on the skin. Amplified electrical signals obtained from
measuring these voltages over time can be fed into electronic devices for further processing. As
speech is produced by the activity of human articulatory muscles, the resulting myoelectric
signal patterns measured at these muscles provides a means of recovering the speech
corresponding to it. Since sEMG relies on muscle activity alone, speech can be recognized even
if produced silently, i.e. without any vocal effort and the signal furthermore cannot be corrupted
or masked by ambient noise transmitted through air.
Reference point
Cheek EMG
signal channel
Lower lip EMG

channel
Chin EMG channel
Positioning of surface electrodes

Commonly three channels are needed for the positioning of electrodes. First from
the cheek about 2.5 cm from the nose, second from the chin and third from the lower lip.
Muscles in these regions are selected for the recording of EMG signals. Also an
additional electrode is attached as a reference point. The reference electrode (at times
called the ground electrode) is necessary for providing a common reference to the
differential input of the preamplifier in the electrode. For this purpose, the reference
electrode should be placed as far away as possible and on electrically neutral tissue (say
over a bony prominence).
Page 9
Seminar Report 2014
CHAPTER 4
WORKING OF EMG
The electromyography sensors attached to the face records the electric signals produced
by the facial muscles and is stored in a computer. In the computer a prerecorded signal of spoken
words is already stored. Then the obtained EMG signal is then compare with prerecorded signal
pattern of spoken words. When there is a match occur that sound is transmitted on to the other
end of the line and person at the other end can hear the voice clearly. Before making a phone
call, the phone is connected via Bluetooth and the sensor is connected through wires to the
computer. Over the years 2007-08, a large data base of EMG signals was collected from 78
speakers in a quiet room. All participants were female adults between 18 and 35 years of age
with normal vocal qualities. Each of the speakers read the same sentences in two different
modes, audible and non-audible mode. In the audible and non-audible part, the specialists
parallely recorded the EMG signal corresponding to each word they spoke. In order to get better
output the speakers read phonetically balanced sentences in a controlled setting rather than
recording conversational unplanned speech.
EMG signals of one sentence, once pronounced audibly (upper row) and once pronounced
silently (lower row)
By comparing and concatenating these two signals, the EMG signal corresponding to
specific word is obtained and speech is produced by specific algorithm.
Page 10
Seminar Report 2014
CHAPTER 4
BLOCK DIAGRAM
ELECTRODES
A pair of electrode is placed on the skin surface to record the muscle activity. Surface
electrode converts the ionic currents generated by the muscle into electronic current and can be
fed to electronic devices for further processing of the signal. The design of the electrode unit is
the most critical aspect of the electronics apparatus which will be used to obtain the signal.
Electrode is made up of the material Ag/AgCl electrode. The inter-electrode distance is 20mm
and the size of the electrode depends on the muscle size, should not exceed 10mm.
AMPLIFIER
In order to eliminate the potentially much greater noise signal from power line sources, a
differential detecting configuration is employed. The signal is detected at two sites; electronic
circuitry subtracts the two signals and then amplifies the difference. As a result, any signal that is
common to both detection sites will be removed and signals that are different at the two sites will
have a differential that will be amplified. Any signal that originates far away from the detection
Page 11
Seminar Report 2014
sites will appear as a common signal, whereas signals in the immediate vicinity of the detection
surfaces will be different and consequently will be amplified. Maximum peak to peak amplitude
of sEMG is 5mV and its gain is about 1000k.
ANTI-ALIASING FILTER
When an analog signal is sampled at rates smaller than twice their higher frequency,
sinusoids with frequency above this threshold superimposed on the low frequency signals. This
phenomenon is called aliasing. Due to this aliasing effect information may be lost and to avoid
this anti-aliasing filter is used. The frequency of the signal is band limited to 15Hz to 400 Hz.
A/D CONVERTER
EMG signals need to be digitized for further processing and data analysis. For that an
analog to digital converter is used.
Page 12
Seminar Report 2014
CHAPTER 5
EMG SIGNAL PROCESSING
5.1 WAVELET ANALYSIS
The time-frequency plane is one of the most fundamental concepts in signal analysis.
Wavelets have been growing in popularity as an alternative to the usual Fourier transform
method. It transforms the signals in both time and frequency domains to determine the dominant
modes of variability and their temporal changes. Removing the interference of random noises
from EMG signals (for example, white Gaussian noise (WGN)) using filtering procedures is
difficult. Wavelet de-noising algorithms can effectively remove these noises and is used for
feature extraction also. Phinyomark provided the basic idea of a wavelet-based de-noising
procedure. It was an easy, simple and inexpensive process. The benefit of using a wavelet basic
function is that it has continuous derivatives, which allows it to decompose a continuous function
more efficiently. It also avoids unwanted signals.
Raw EMG signal de-noised by wavelet function
Page 13
Seminar Report 2014
CHAPTER 6
NOISE SOURCES IN EMG SIGNALS
Whenever an EMG signal is being recorded from the muscle, various types of noises
contaminate it. Therefore, analyzing and classifying the EMG signals is very difficult because of
the complicated pattern of the EMG. It is very difficult for the noise to be removed clearly.
Therefore, that EMG signal is processed and analyzed to get the required information. The main
challenges in analyzing the EMG signals are explained below.
Inherent Noise in the Electrode
All types of electronic equipment generate electrical noise, otherwise known as inherent
noise. This noise has frequency components that range from 0 Hz to several thousand Hz. When
the electrode size enlarges, the impedance decreases. However, electrode size should not be very
large. On the other hand, high electrode impendence effectively reduces the signal quality and
gives low signal-to-noise ratio. Therefore, both parameters should be taken into consideration.
Researchers are allowed to use high electrode impedances for experiments in which statistical
power is high or in which large numbers of electrodes are necessary, but tend to switch to low
electrode impedances for experiments in which statistical power would otherwise be too low.
This noise can be eliminated by using intelligent circuit design and high quality instruments.
Movement Artifact
Movement of the cable connecting the electrode to the amplifier and the interface
between the detection surface of the electrode and the skin creates motion artifacts. Muscle fibers
generate electric activity whenever muscles are active. EMG signals are recorded by placing
electrodes close to the muscle groups. When the muscle is activated, the length of the muscle
decreases and the muscle, skin and electrodes move with respect to one another. At that time, the
electrodes will show some movement artifacts. The frequency range of the motion noise is
usually 110 Hz and has a voltage comparable to the amplitude of the EMG. Recessed electrodes
can remove the movement artifact significantly, in which a conductive gel layer is used between
the skin surface and the electrode-electrolyte interface.
Page 14
Seminar Report 2014
Electromagnetic Noise
The human body behaves like an antennathe surface of the body is continuously
inundated with electric and magnetic radiation, which is the source of electromagnetic noise.
Electromagnetic sources from the environment superimpose the unwanted signal, or cancel the
signal being recorded from a muscle. The amplitude of the ambient noise (electromagnetic
radiation) is sometimes one to three times greater than the EMG signal of interest. This can be
removed by suitable filtering technique.
Cross Talk
An undesired EMG signal from a muscle group that is not commonly monitored is called
Crosstalk. Crosstalk contaminates the signal and can cause an incorrect interpretation of the
signal information. Crosstalk depends on the many physiological parameters and can be
minimized by choosing electrode size and inter-electrode distances (typically 12 cm or the
radius of the electrode) carefully.
Page 15
Seminar Report 2014
CHAPTER 7
APPLICATONS
The Technology opens up a host of application such as mentioned below:
As we know in space there is no medium for sound to travel therefore this technology can
be best utilized by astronauts.
Helping people who have lost their voice due to illness or accident.
We can make silent calls even if we are standing in a crowded place.
Allow people to make silent calls without bothering others.
Telling a trusted friend your PIN number over the phone without anyone eavesdropping assuming no lip-readers are around.
Silent Sound Techniques is applied in Military for communicating secret/confidential

matters to others.
Since the electrical signals are universal they can be translated into any language. Native
speakers can translate it before sending it to the other side. Hence it can be converted into
any language of choice currently being German, English & French.
Page 16
Seminar Report 2014
CHAPTER 8
ADVANATAGES
Very useful for those people who lost their voice and has been rendered mute due to
accident.
Very good technology for noise cancellation technique.
Helps in making phone calls in noisy environment.
Very useful for sharing confidential information like secret PIN number on phone at
public place.
Very useful for astronaut.
Page 17
Seminar Report 2014
CHAPTER 9
DISADVANTAGES
Translation into majority of languages but for languages such as Chinese different tone
holds different meaning, facial movements being the same. Hence this technology is
difficult to apply in such situations.
It is difficult to recognize who is on the other end of the phone call.
Even differentiating between people and emotions cannot be done. This means we will
always feel you are talking to a robot.
Electrodes are hanging all around our face which is quite impractical to make it usable.
Page 18

Seminar Report 2014 Silent Sound Technology

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Seminar Report 2014 Silent Sound Technology

Hochgeladen von

Copyright:

Verfügbare Formate

Seminar Report 2014

Silent Sound Technology

Dept. of ECE, T.K.M Institute Of Technology, Kollam