R a n d a l l M o r t o n BBus BComp BPsych |
Body | Brain | CIA and AMF | Communication | Cyborgs | Ear | Eyes | Functional Electrical Stimulation | Hacking
History | Lifecycle | Media | Mental Disorders | Microchip | Monitoring
Nerves | Nerve Disorders | Now | Power | Psychology Report | Paraplegic |
Radio Frequency Identification (RFID) tags | Radio Receiver
Radio | Scan for implant | Synapse | Speech and thought | Voice | Search
Speech and thought
http://www.nasa.gov/home/hqnews/2004/mar/HQ_04093_subvocal_speech.html
http://www.nasa.gov/centers/ames/news/releases/2004/subvocal/subvocal.html
|
NASA scientists have begun to computerize human, silent
reading using nerve signals in the throat that control speech. In preliminary
experiments, NASA scientists found that small, button-sized sensors, stuck
under the chin and on either side of the ¡¥Adam¡¦s apple,¡¦ could gather nerve
signals, send them to a processor and then to a computer program that
translates them into words. |
|
|
|
NASA scientists have begun to computerize human, silent
reading using nerve signals in the throat that control speech. In preliminary
experiments, NASA scientists found that small, button-sized sensors, stuck
under the chin and on either side of the ¡¥Adam¡¦s apple,¡¦ could gather nerve
signals, send them to a processor and then to a computer program that
translates them into words. |
|
|
To learn more about what is in the patterns of the nerve
signals that control vocal chords, muscles and tongue position, NASA Ames
scientists are studying the complex nerve signal patterns. "We use an
amplifier to strengthen the electrical nerve signals. These are processed to
remove noise, and then we process them to see useful parts of the signals to
show one word from another," Jorgensen said. |
|
|
In their first experiment, scientists ¡¥trained¡¦ special
software to recognize six words and 10 digits that the researchers ¡¥repeated¡¦
subvocally. Initial word recognition results were an average of 92 percent
accurate. The first sub-vocal words the system ¡¥learned¡¦ were ¡¥stop,¡¦ ¡¥go,¡¦
¡¥left,¡¦ ¡¥right,¡¦ ¡¥alpha¡¦ and ¡¥omega¡¦ and the digits ¡¥zero¡¦ through ¡¥nine.¡¦
Silently speaking these words, scientists conducted simple searches on the
Internet by using a number chart that represents the alphabet to control a
Web browser program. Please credit photo to NASA Ames Research Center,
Dominic Hart.
|
|
|
A second demonstration will be to control a mechanical
device using a simple set of commands, according to Jorgensen. His team is
planning tests with a simulated Mars rover. "We can have the model rover
go left or right using silently ¡¥spoken¡¦ words," Jorgensen said.
|
|
|
A second demonstration will be to control a mechanical
device using a simple set of commands, according to Jorgensen. His team is
planning tests with a simulated Mars rover. "We can have the model rover
go left or right using silently ¡¥spoken¡¦ words," Jorgensen said. |
|
|
People in noisy conditions could use the system when
privacy is needed, such as during telephone conversations on buses or trains,
according to scientists. |
http://sigint.wordpress.com/2008/02/09/subvocal-recognition-using-electromyography/
NASA Research: Reading Thoughts Using Electromyography
Victims of mind control report having their thoughts read especially those that are verbalized internally otherwise know as subvocal speech. Subvocal speech happens also when we read although some ¡§speed reading¡¨ techniques try to get you to stop doing this because it slows the reading process down. It has long been known that during subvocal speech electrical signals are still sent to those muscles of the face and throat that participate in ordinary speech even though no sounds are generated. These signals can be detected in the form of electromyograms (EMG) using electrodes placed in the face and neck area.
In the previous article we saw how Lawrence Pinneo found some success in using electromyograms and electroencephalograms to detect thoughts in the form of subvocal speech. That was thirty years ago, this article looks at current day attempts to do this through the work of Charles Jorgenson at NASA. Jorgensen works at NASA¡¦s Ames Research Center where they have been investigating alternative methods of communication and control in hostile environments where normal methods are not always possible. Examples include astronauts, fighter pilots and rescue workers.
In contrast to Pinneo, Jorgensen placed electrodes in the neck region only. In the first stage of processing signals from the electrodes were sampled at up to 10,000 times per second and run through a 60 HZ notch filter to exclude line interference and band pass filters to remove anything outside the 30 ¡V 500 HZ range.
The next stage in processing marks a major change from Pinneo¡¦s work in which the data is transformed from the time domain to the frequency domain. Jorgensen experimented with a number of transforms including Fourier and Wavelet transforms to do this and seemed to settle on a quad tree wavelet transform.
During the final phase features extracted from the transforms were input into different machine learning algorithms to train the system to pattern match the words or phonemes being examined. Jorgensen experimented with a number of techniques including neural networks and support vector machines. Once the system had been trained it was then used to attempt to match against new signals. In a range of different applications was able to achieve around 74% success for small vocabulary sets of up to 15 words.
The results continue to prove the feasibility of thought reading using subvocal recognition. However given the vast increase in computing power plus the many advancemnets in signal processing and pattern recognition since Pinneo¡¦s time these results are somewhat disappointing although I should point out that nothing has been published by Jorgensen in 2-3 years. Not sure if we should read anything into that.
Despite this, some companies are looking to commercialize this technology. For example NTT DoCoMo are working on a subvocal mobile phone with the idea that people can answer their phone without annoying those around them such as in a movie theater. NASA are also working with QUASAR corporation to develop better sensors. In April 2006 Forbes magazine published an article on Jorgensen¡¦s work entitled The Silent Speaker:
Jorgensen sees the day when electromagnetic sensors will be woven into the fibers of turtlenecks or rescue workers¡¦ outfits. ¡§As long as people have had machines and tools, they¡¦ve been dependent on the physicality of the body,¡¨ Jorgensen says. ¡§Separate those control activities from the body and it opens a whole new generation of interface design.¡¨
QUASAR recently announced that it is working with NASA Ames Research Center to develop a hands-free UGV (unmanned ground vehicle) control system based on subvocal speech and forearm EMG. The purpose of the system is to allow soldiers to control the devices without having to set down their weapons or other equipment.
In December 2001 NASA¡¦s Ames Research Center made a presentation to NorthWest Airlines in which they stated that they were working with a commercial partner to develop neuro-electric sensors to remotely monitor the EEG and ECG of passengers at airport security. This information was leaked to the Washington Times which published an article about it. The article drew denials from NASA and the Washington Times no longer has the article up on their site. The Electronic Privacy Information Center obtained details of the presentation under the Freedom of Information Act which can be seen here.
References
- Techniques of EMG signal analysis: detection, processing, classification and applications, Raez, Hussain and Mohd-Yasin 2006
- Small Vocabulary Recognition Using Surface Electromyography in an Acoustically Harsh Environment, Bradley J. Betts, Charles Jorgensen, 2005
- Web Browser Control Using EMG Based Sub Vocal Speech Recognition, Chuck Jorgensen, Kim Binsted, 2004
- NASA Develops System To Computerize Silent, ¡§Subvocal Speech¡¨, NASA News,
- The Silent Speaker, David Armstrong, Forbes Magazine, April 10 2006
http://thefutureofthings.com/articles.php?itemId=28/58/
http://www.nasa.gov/centers/ames/research/technology-onepagers/human_senses.html
http://www.lti.cs.cmu.edu/Research/Thesis/Jou_SzuChen_phd_thesis.pdf
http://www.newscientist.com/article/dn4795-nasa-develops-mindreading-system.html
NASA develops 'mind-reading' system
- 15:35 22 March 2004 by Maggie McKee
A computer program which can read silently spoken words by analysing nerve signals in our mouths and throats, has been developed by NASA.
Preliminary results show that using button-sized sensors, which attach under the chin and on the side of the Adam's apple, it is possible to pick up and recognise nerve signals and patterns from the tongue and vocal cords that correspond to specific words.
"Biological signals arise when reading or speaking to oneself with or without actual lip or facial movement," says Chuck Jorgensen, a neuroengineer at NASA's Ames Research Center in Moffett Field, California, in charge of the research. Just the slightest movements in the voice box and tongue is all it needs to work, he says.
The sensors have already been used to do simple web searches and may one day help space-walking astronauts and people who cannot talk. The system could send commands to rovers on other planets, help injured astronauts control machines, or aid disabled people.
In everyday life, they could even be used to communicate on the sly - people could use them on crowded buses without being overheard, say the NASA scientists.
Web search
For the first test of the sensors, scientists trained the software program to recognise six words - including "go", "left" and "right" - and 10 numbers. Participants hooked up to the sensors silently said the words to themselves and the software correctly picked up the signals 92 per cent of the time.
Then researchers put the letters of the alphabet into a matrix with each column and row labelled with a single-digit number. In that way, each letter was represented by a unique pair of number co-ordinates. These were used to silently spell "NASA" into a web search engine using the program.
"This proved we could browse the web without touching a keyboard," says Jorgensen.
Noisy settings
Phil Green, a computer scientist focusing on speech and hearing at the University of Sheffield, UK, called the research "interesting and novel" on hearing the news. "If you're not actually speaking but just thinking about speaking then at least some of the messages still get sent from the brain to the vocal tract," he says.
But he cautions the preliminary tests may have been successful because of the short lengths of the words and suggests the test be repeated on many different people to test the sensors work on everyone.
The initial success "doesn't mean it will scale up", he told New Scientist. "Small-vocabulary, isolated word recognition is a quite different problem than conversational speech, not just in scale but in kind."
He says conventional voice-recognition technology is more powerful than the apparent results of these sensors, and that "the obvious thing is to couple this with acoustics" to enhance communication in noisy settings.
The NASA team is now working on sensors that will detect signals through clothing.