Debug: Database connection successful Subvocalization Technology and Practice / Science, Technology, and Astronomy / New Mars Forums

New Mars Forums

Official discussion forum of The Mars Society and MarsNews.com

You are not logged in.

Announcement

Announcement: This forum is accepting new registrations via email. Please see Recruiting Topic for additional information. Write newmarsmember[at_symbol]gmail.com.

#1 Today 06:14:41

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 23,322

Subvocalization Technology and Practice

This topic is about a procedure I've been thinking about for many decades.

Thanks to the arrival of powerful AI tools, I now have access to information that could lead to a practical device.

I'll put a conversation with Gemini into Post #3

This topic is available for all NewMars members to contribute links, images or text that might help to realize a practical device to collect data from inside the human body for communication or for machine control.

(th)

Offline

Like button can go here

#2 Today 06:15:49

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 23,322

Re: Subvocalization Technology and Practice

This post is reserved for an index to posts that might be contributed by NewMars members.

Index:
Post #3: Conversation with Gemini about Subvocalization Technology, including ongoing research

(th)

Offline

Like button can go here

#3 Today 06:16:58

tahanson43206
Moderator
Registered: 2018-04-27
Posts: 23,322

Re: Subvocalization Technology and Practice

This post records a conversation with Google's Gemini, about subvocalization detection technology, including research and practical applications if any.

Conversation with Gemini
Sensor System detect subvocalization - Hello Again! And thank you for your help with many previous requests! I'm hoping you find this one interesting as well as challenging. I am hoping you can find reports of researchers able to detect electronic activity in the human body that occurs when a human being thinks about words or tones. While lying quietly in bed, my brain is hard at work (upon occasion) and I can subvocalize the Morse Code. If I had a detector of some kind, I could pick up the subvocalized code elements and translate them into audio which could then be broadcast over an ordinary speaker. This system could also help persons who are suffering paralysis (hopefully temporary)l It could also be used for equipment operation in some situations, but I haven't thought of specific examples. Please search the files available to you to see if there is anything like this on record.
***
This is a fascinating area of research, and you are spot on—your "detector" concept is not only theoretically possible but is actively being developed. The phenomenon you are describing is called Subvocalization (or "silent speech"), and the specific electronic activity you are asking about is detected using Electromyography (EMG).

While you are lying in bed "subvocalizing" Morse Code, your brain is sending neuromuscular signals to your vocal cords, tongue, and jaw. Even if you don't make a sound or visibly move your mouth, these tiny electrical impulses (microvolts) are firing.

Here is a breakdown of the research and systems that match your description.

1. The Technology: "Reading" the Muscles
Researchers have found that when you think about saying a word, your brain sends nearly the same signals to your speech muscles as it would if you were speaking aloud.

EMG Sensors: These are sticky pads (electrodes) placed on the throat (larynx), under the chin, or on the jaw. They detect the electrical potential generated by muscle cells.

The "Morse Code" Advantage: Your idea to use Morse Code is brilliant because it simplifies the problem. Detecting complex words (like "Hello") requires distinguishing between dozens of complex muscle patterns. Detecting a "dit" (short signal) and a "dah" (long signal) is a simple binary distinction. This would be much easier to build than a full speech recognizer.

2. Real-World Examples & Research
There are specific projects that operate exactly as you described:

MIT's AlterEgo (The "Mind-Reading" Headset)

What it is: A wearable headset developed at the MIT Media Lab (led by Arnav Kapur). It hooks over the ear and rests along the jawline.

How it works: It uses sensors to detect neuromuscular signals from the jaw and face when the user subvocalizes.

Capabilities: It can translate silent speech into computer commands. Users can "ask" the internet a question silently and hear the answer through bone-conduction headphones (which vibrate the skull to create sound), keeping the entire loop silent.

Relevance to you: This is essentially the high-tech version of the "detector" you envisioned.

NASA's "Subvocal Speech" Systems

The Researcher: Dr. Chuck Jorgensen at NASA’s Ames Research Center.

The Experiment: NASA developed sensors placed on the throat to control a Mars rover simulator. The operator simply subvocalized commands like "stop," "go," or "left," and the rover obeyed.

Why they did it: In space, astronauts often work in loud environments or thick suits where microphones fail. Subvocalization provides a clean, noise-free signal.

3. Applications for Paralysis & Equipment Operation
You mentioned this could help people with paralysis; this is one of the primary drivers of this research.

ALS & Stroke Patients: For patients who have lost the physical ability to speak but retain cognitive function (often called "Locked-in Syndrome"), the neural pathways to the vocal cords may still be active even if the muscles are too weak to produce sound. An amplified "subvocal detector" can act as a prosthetic voice.

Silent Communication: Military and special operations teams are interested in this for "silent radio," allowing soldiers to communicate without making a sound that could give away their position.

4. How you could build this (The "Hobbyist" Approach)
Since you are interested in a "detector of some kind," you can actually experiment with this using consumer-grade electronics:

OpenBCI: This is an open-source platform that makes biosensing boards. You could use an EMG board (like the Cyton) with surface electrodes placed on your throat or jaw.

The Morse Code Experiment:

Place EMG electrodes on your jaw (masseter muscle) or throat.

Lie in bed and subvocalize a "dit" (short tension) and a "dah" (long tension).

Watch the signal spike on a computer screen.

With simple software (like Python), you could trigger a computer to play a "beep" whenever the signal crosses a certain threshold, effectively creating the broadcast system you imagined.

Summary
The system you imagined is Subvocal Recognition (SVR). It picks up electromyographic (EMG) signals from the larynx and articulatory muscles. It is currently used in high-end labs (MIT, NASA) but is accessible enough that a dedicated hobbyist could build a prototype for Morse Code translation.

Next Step: Would you like me to find a simple wiring diagram or a tutorial on how to set up a basic EMG sensor with a microcontroller (like an Arduino) to detect these muscle spikes?
***
Thank you ** very ** much for your inspiring report, and for the hard work that went into it! I will publish your report (with all due credit) on a small international forum. If there is interest, I will return to consult with you again.
***
You are most welcome! I am delighted to hear that you found the report inspiring.

It is a fascinating subject—the intersection of human biology and electronic interfaces is one of the most exciting frontiers in technology, especially when applied to assistive devices like the one you envisioned.

Good luck with your post on the international forum. It is an honor to have contributed to the discussion. I look forward to hearing what the members think!

Would you like me to generate a brief, bulleted summary or "abstract" of the key findings (defining terms like EMG and Subvocalization) to make it easier for you to format your post?

(th)

Offline

Like button can go here

Board footer

Powered by FluxBB