0% found this document useful (0 votes)
9 views6 pages

Matrix Coding for Subvocal Speech

Researchers at NASA are developing a technology called subvocal speech that allows communication without audible speech by decoding electrical signals produced during silent thought. To do this, electrodes are placed on the larynx to record sound signals, which are assigned number pairs in a matrix code since the technology cannot recognize letters. In a test, NASA researchers were able to direct a small Mars rover to different locations using only thought commands by thinking, not speaking, commands like "right" and "left." The matrix assigns numbers to letters, like 1,1 for A and 1,2 for B, to decode thoughts into words without recognizing the alphabet.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
9 views6 pages

Matrix Coding for Subvocal Speech

Researchers at NASA are developing a technology called subvocal speech that allows communication without audible speech by decoding electrical signals produced during silent thought. To do this, electrodes are placed on the larynx to record sound signals, which are assigned number pairs in a matrix code since the technology cannot recognize letters. In a test, NASA researchers were able to direct a small Mars rover to different locations using only thought commands by thinking, not speaking, commands like "right" and "left." The matrix assigns numbers to letters, like 1,1 for A and 1,2 for B, to decode thoughts into words without recognizing the alphabet.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

The Matrix Decoded

Teacher Information

Background Information
Have you ever noticed that when you really think about what you are reading, your lips
sometimes move and you silently “speak” the words? There’s a reason for that. The
brain sends out signals that connect what we think and what we say. We usually don’t
“read along with our lips” because it slows us down when we read. Still, it is a natural
connection. Researchers at NASA are learning how that thought-speech connection
can be used to help us communicate even better.

To make this technology work, researchers place electrodes on the larynx of the user.
The electrodes record sound signals. The device doesn’t record words, though, it
records sounds. And each sound is assigned a pair of numbers. Researchers use
numbers to code the alphabet into a matrix. For example 1,1 is used to record the
letter a. The letter b is shown as 1,2 and so on. This coding is used because
subvocal speech technology doesn’t recognize the alphabet. In a test, NASA
researchers used the new technology to control a small Mars rover. They were able
to direct the rover to different locations without any audible sounds. Commands for
the rover were thought, but not spoken. Using the new technology, the rover moved in
the correct direction.

In this lesson, students will learn how the researchers use a matrix to assign numbers
to each sound. They will then use this code to decode a series of words and sayings.

Key Concepts and Student Information


• Matrix – a table of intersecting rows and columns
• Subvocal speech – communicating without actually talking

Materials
NASAexplores 5-8 article, “Do You Hear What I’m Saying?” (1 copy per student)
Student Sheets (1 copy per student)

Procedure
1. Read the NASAexplores 5-8 article, “Do You Hear What I’m Saying?” Discuss.
2. Revisit the paragraph that discusses how researchers use a matrix to assign
number pairs to different letters of the alphabet.
3. Explain that a matrix is a grid of intersecting rows and columns. Distribute the
Student Sheets to give students an example. Be sure to point out that letter a is
represented by the pair 1,1 just like in the article.
4. Go over Student Sheet instructions. Answer questions before students begin.

Enrichment Activities
• Discuss other uses for the matrix coding system.
• Have students come up with their own puzzles to be decoded.
• Discuss possible uses for subvocal speech technology.

Answer to Student Sheet Message: In the future, subvocal technology could be


used by SCUBA divers to speak to each other while under water.
The Matrix Decoded

Do You Hear What I’m Saying?


NASAexplores article, August 25, 2004
[Link]

Have you ever noticed that when you


concentrate on what you’re reading,
your lips sometimes move and
silently speak the words? There’s a
reason for that. The brain sends out
electrical signals that connect
thinking and speech. We often avoid
“reading with our lips” because it
slows down the process, but it’s a
natural connection. Researchers at
NASA are learning how that thought-
speech connection can be used for
better communication.

Imagine being able to speak to someone without ever making a sound. That’s the idea
behind the subvocal speech project at NASA’s Ames Research Center. Researchers
have found that when people think of something to say, the brain sends out faint
electrical signals with that same message—even if the person never speaks the words
out loud. When researchers magnify those electrical signals and translate them into
real words, people can think a message and the message will be turned into words.
This is a form of voice recognition. It sends a message while cutting out the breath
and sound vibration parts of speech. This technology is called subvocal speech.
Subvocal means “below the spoken voice,” referring to the nerve signals to the
muscles that we use when we produce sounds.

Yelling is one way of sending a message


to others. Why would it be useful to speak
without using your voice? “People
working in high-noise areas, such as
construction sites or airports, could use
subvocal communication to send
commands and instruction for tasks when
they otherwise couldn’t be heard,” says
Chuck Jorgensen, chief scientist for
neuroengineering at NASA Ames.
“People who have suffered damage to
their vocal cords could communicate by
thinking, rather than speaking, their thoughts. Space explorers who may get injured
while on Mars could use subvocal speech to call for help. And when sitting in a
crowded room, people could ask each other a private question without bothering the
rest of the people.” Passwords could be entered without anyone else hearing the
word. Deep sea divers could communicate to diving partners in an environment where
words are impossible. Telephone calls could be conducted silently.
The Matrix Decoded

Do You Hear What I’m Saying? (Continued)

Don’t get the wrong idea about subvocal speech, though. It’s not mind reading. It’s just
like typing a word on a computer keyboard, but the message is sent by thinking the
idea rather than by typing the words. Subvocal speech is totally voluntary; nobody
forces anyone to think on demand.

To make this technology work, small


electrodes are placed on the throat near the
larynx. The signal that is created by thinking
is captured and recorded. The list of
recorded nerve patterns creates a type of
dictionary that a computer remembers. So
far, the system is able to recognize a few
words, and only words spoken by a
particular voice. As the system improves, it
will be able to recognize more words and
more speech patterns.

Researchers use numbers to code the alphabet into a matrix, so that 1,1 is the letter
a, and 1,2 is b, and so on. They do this because subvocal speech technology cannot
yet recognize the alphabet. In a demonstration, NASA researchers were able to
control a small Mars rover by directing the rover without any audible sound to go to
different locations. Commands such as right and left were thought,
but not spoken, and the rover responded.

Everyone's voice creates a unique [Link] electronic


“signature” put out by your muscles is unique, just like fingerprints,
Jorgensen says. Identifying the various signals that translate into
different sounds and words is the challenge the researchers face
now. Not only is every word different, but each person pronounces
those words a bit differently, too. Regional accents and individual
pronunciation create big differences in the signal.

Because English has so many different ways to pronounce the


same letters, the researchers decided to use Latin when they first
experimented with subvocal speech. English has
diphthongs—combinations of vowels that create unique sounds,
such as the word “voice”—and depending on how the letters are
arranged, letters can have several different pronunciations. “Nasal
languages, such as English and French, turned out to be much harder to develop into
subvocal speech,” says Jorgensen. “Languages with hard, guttural sounds, like
German and Japanese, are easier. They are more consistent in how each letter is
pronounced. And, Latin was the easiest tool for pulling out sounds and making
consistent pronunciation. So, there’s one more reason to know another language!”
The Matrix Decoded

Student Sheets

Objective
To understand how to decode a message using a matrix.

Materials
Matrix Decoded Worksheet
Do You Hear What I’m Saying article

Procedure
1. Read the “Do You Hear What I’m Saying?” article from NASAexplores.
2. Practice identifying letters using the matrix code on the Matrix Decoded student
worksheet
3. When you have finished the Decoding Practice and Coding Practice, go on to
decode the message on worksheet #2.
4. After decoding the message, create a message of your own, and code it according
to the matrix.
5. Have someone else in your group decode your message.
The Matrix Decoded

Student Worksheet

Subvocal speech technology cannot recognize the alphabet yet, so researchers use
numbers to code the alphabet into a matrix, so that (1,1) is the letter A, and (1,2) is B, and
so on.

The box below is called a matrix; it is a grid with intersecting rows and columns.
Researchers at NASA use a matrix, like the one below, to assign a letter to a pair of
numbers. For example, the pair of numbers (3,2) is found on the matrix by finding the
number 3 in the top row and seeing where it intersects with the row containing the
number two in the left vertical column.

1 2 3 4 5
1 A B C D E
2 F G H I J
3 K L M N O
4 P Q R S T
5 U V W X Y
6 Z

In the matrix above the horizontal and vertical rows intersect at the letter H. Therefore,
whenever the pair of numbers (3,2) is seen, the computer will interpret the code as the
letter H. Some other examples are:
(4,4) = S
(1,5) = U

DECODING PRACTICE:

( 2, 4) = _________ (5, 3) = __________ (3, 5) = __________

You can also use the matrix backward to create your own secret message! In order to do
this first, create your own secret message. Then, for each letter in the message find the
corresponding pair of numbers. For example, if your message you wanted code is the
word NASA, then your secret code would be:
NASA = (4,3) (1,1) (4,4) (1,1)

CODING PRACTICE:

R = ( ____, ____) C = ( ____, ____) I = ( ____, ____)


The Matrix Decoded

Student Worksheet #2

Try out your new decoding skills, use this matrix to decode the secret messages below.

1 2 3 4 5
1 A B C D E
2 F G H I J
3 K L M N O
4 P Q R S T
5 U V W X Y
6 Z

___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___,
(4,2) (4,3) (5,4) (3,2) (5,1) (1,2) (1,5) (5,4) (1,5) (3,4) (5,1)

___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___
(4,4) (1,5) (2,1) (2,5) (5,3) (3,1) (1,1) (2,3) (5,4) (5,1) (3,1) (3,2) (4,3) (5,3) (2,3) (5,3) (2,2) (5,5)

___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___
(3,1) (5,3) (1,5) (2,3) (4,1) (2,1) (5,1) (1,5) (4,4) (5,1) (4,1) (2,1) (5,5)

___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___
(4,4) (3,1) (1,5) (2,1) (1,1) (4,1) (4,2) (2,5) (5,1) (3,4) (4,4) (5,4) (5,3)

___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___
(4,4) (1,4) (5,1) (1,1) (1,3) (5,4) (5,3) (5,1) (1,1) (3,1) (3,2) (5,3) (5,4) (3,2) (5,1) (3,4)

___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___.
(3,5) (3,2) (4,2) (2,3) (5,1) (1,5) (4,3) (4,1) (5,1) (3,4) (3,5) (1,1) (5,4) (5,1) (3,4)

Use the space below to write out your own short secret message

____________________________________________________________________

Now, using the matrix above code your secret message into pairs of numbers.

Common questions

Powered by AI

Subvocal speech technology operates by placing electrodes near the user's larynx to capture faint electrical signals generated by the brain when thinking of words to speak, even if no sound is produced. These signals are captured, amplified, and translated into words using a type of voice recognition system. The matrix coding system comes into play because subvocal technology cannot yet recognize the alphabet and words outright. Researchers use a coded matrix to assign numbers to letters (for example, 1,1 represents 'a', 1,2 represents 'b'), enabling the computer to interpret the electrical signals as specific letters and words .

Matrices facilitate the coding and decoding processes within subvocal speech technology by providing a structured system to convert electrical signals into recognizable symbols. Each letter of the alphabet is represented by a unique coordinate pair in a matrix, such as (1,1) for 'a' and (1,2) for 'b'. This numerical representation allows the computer to interpret the signals produced by subvocalizations as specific letters or words, bridging the gap between neural signals and device-understandable data .

Subvocal speech technology has several potential applications, particularly in environments where traditional speech is ineffective. People working in noisy conditions, such as at construction sites or airports, can use this technology for clear communication. It can also help individuals who have lost vocal cord function to communicate by thinking their words. Space explorers might use it in emergencies for silent communication. It also provides a means for private conversations in crowded settings and secure communication for entering passwords silently. Ocean divers may use it to communicate underwater, and phone calls could be conducted in silence .

Subvocal speech technology fundamentally challenges traditional communication paradigms by allowing individuals to 'speak' without vocalizing words. It replaces voice imposition by capturing neural signals associated with speech, effectively enabling silent communication. This paradigm shift broadens communication possibilities in environments where vocal speech is impractical or impossible, such as in noise-dominated areas or under water, and facilitates private exchanges in public spaces. Additionally, it offers a form of interaction for those with vocal impairments, redefining accessibility in communication .

Educational activities derived from understanding the matrix system include decoding exercises where students practice converting letter codes into words and create their own puzzles for peers to solve. These activities improve comprehension of grid systems and symbolic representation, fostering analytical skills. Enrichment activities could include discussions on alternative applications of matrix coding and the creation of personalized messages using the matrix. These exercises benefit students by enhancing problem-solving abilities, encouraging creativity, and fostering an understanding of language and coding systems .

Researchers overcome the limitations of complex English speech patterns by initially using Latin, a language with more consistent pronunciation due to its lack of diphthongs and varying vowel sounds. By focusing on Latin, researchers can more easily differentiate between sound patterns and develop a foundational understanding of subvocal signals in a controlled linguistic environment. This approach helps create a base upon which more complex languages can be added, ensuring that foundational principles are clear before tackling the diverse challenges of English and other nasal languages .

Researchers face the challenge of accurately identifying various electrical signals that correspond to different sounds and words, which are unique to each individual's voice. This is further complicated by regional accents and individual pronunciation differences, making the task of recognition difficult. The complexity of the English language, with its diphthongs and varying pronunciations, adds to the difficulty. Researchers initially used Latin for experiments due to its consistent pronunciation, finding it easier to decode compared to nasal languages like English and French. Guttural languages such as German and Japanese are also easier to work with due to consistent pronunciation .

Using a consistent language like Latin provides clear advantages in developing subvocal speech technology because it lacks the complexities found in English, such as diphthongs and variable pronunciation. It offers stability and predictability, allowing researchers to reliably interpret and build vocabularies based on fixed pronunciations. This approach lays a systematic foundation before tackling more complex languages. It thus accelerates research by reducing variability, enabling researchers to refine recognition algorithms that can later adapt to varied linguistic patterns when applied to other languages .

The uniqueness of an individual's electrical signals poses a significant challenge because each person's voice generates distinct neuro-electrical patterns, analogous to fingerprints. This means that the technology must be finely tuned to the specific signal patterns of each user, complicating the creation of a universal system. Additionally, variances in accent and pronunciation play a role, necessitating potentially extensive individual calibrations and signal differentiations before the technology can be widely implemented .

Subvocal speech technology is similar to typing on a computer in that both involve converting thoughts into digital information that can be communicated to others. However, instead of physically typing words, subvocal technology uses neural signals associated with speech to perform this task silently. The difference lies in the input method; typing requires manual keyboard interaction, whereas subvocal technology requires the translation of thought-generated electrical signals into recognizable words without vocal sound production, providing a layer of intangibility and privacy .

You might also like