Signals in the bird’s brain were read by scientists as a breakthrough event that could help develop prosthetic limbs for humans who have lost their ability to speak.
In the study, silicone implants recorded firing of brain cells when an adult male zebra finch passed through the complete repertoire of songs.
By supplying brain signals via artificial intelligence, college teams California San Diego predicts what the bird will sing next.
Breakthrough advances open the door to new devices that can be used to turn the thoughts of those who cannot speak into real spoken language for the first time.
Today’s state-of-the-art implants allow users to generate text at speeds of about 20 words per minute, but this technique allows for a completely natural “new voice.”
Research co-author Timothy Gentner said he imagined prosthetic limbs for voiceless people, which allowed them to communicate naturally by voice.
Diagram of the experimental workflow. When a male zebra finch sings his song, which consists of a sequence of “1, 2, 3”, he thinks about the next syllable to sing (“4”).
In the study, silicone implants recorded firing of brain cells when an adult male zebra finch passed through the complete repertoire of songs.Stock image
How they predicted bird speech
Researchers implanted a silicon electrode in the brain of an adult male zebra finch to record neural activity while the bird was singing.
They studied a specific set of electrical signals called the local electric field potential.
These signals were recorded in the parts of the brain needed to learn and generate songs.
Known as “local field potential,” they found that they were translated into specific syllables of bird songs.
It then predicts when syllables will occur between songs.
Lead author Darryl Brown, a PhD student in computer engineering, said research with bird brains “sets the stage for a bigger goal” of giving voice to the voiceless.
“We are studying bird calls in a way that helps us take a step closer to engineering brain-machine interfaces for vocalization and communication.”
Bird barks and human speech are both learned behaviors and share many characteristics, including the fact that they are more complex than the barks of other animals.
The team used signals from the bird’s brain to focus on a series of electrical signals called “local electric field potentials.”
These are necessary to learn and produce songs.
They have already been thoroughly studied in humans and used to predict the vocal behavior of zebra finch.
Professor Vikash Gilja, co-leader of the project, said:
‘This treatise shows that there are many similarities in this type of signaling between zebra finch and humans, and other primates.
“These signals can be used to begin deciphering the brain’s intent to produce speech.”
Various functions are translated into specific “syllables” of bird songs, indicating when they occur and enabling predictive algorithms.
“With this system, we can faithfully predict the onset of vocalization of a songbird. We can predict in what sequence the bird will sing and when it will sing,” Brown explains.
They even expected variations in the sequence of the songs, down to the syllables.
Professor Timothy Gentner, co-leader of the project, said: ‘In the long run, we would like to use the detailed knowledge gained from the songbird brain to develop a communication prosthesis that can improve the quality of life of humans suffering from various diseases and disorders. ‘
Elon Musk Neuralink lets monkeys play Pong with their hearts
Elon MuskNeuralink unveiled the latest brain implants by letting monkeys play Pong with their hearts.
The brain computer interface was embedded in a 9-year-old macaque monkey called a pager.
A device in his brain recorded information about neurons that fired while he was playing the game.
Musk said on Twitter:
Last month, a tech tycoon told Twitter users that he had been approved to work with the US Food and Drug Administration to initiate human clinical trials.
For example, you can build on a set of four iterations. From time to time, change to 5 or 3. Changes in the signal revealed them.
If the bird song is based on a repeating set of syllables “1, 2, 3, 4”, the sequence will change to “1, 2, 3, 4, 5” or “1”. there is. , 2, 3. “
The characteristics of the local field potential reveal these changes, the researchers found.
“These morphological changes are important for testing fictitious artificial speech, because humans don’t just repeat a sentence over and over,” says Professor Gilja.
“It’s exciting to find similarities in the brain signals recorded and documented in human physiology studies with those in songbird studies.”
Conditions associated with loss of voice or language function range from head trauma to dementia and brain tumors.
Professor Timothy Gentner, co-leader of the project, said: ‘In the long run, we would like to use the detailed knowledge gained from the songbird brain to develop a communication prosthesis that can improve the quality of life of humans suffering from various diseases and disorders. .. “
SpaceX founder Elon Musk and Facebook CEO Mark Zuckerberg are currently working on a brain-reading device that allows thought to send text.
Research PLoS Computational Biology..
How to Learn Artificial Intelligence Using Neural Networks
AI systems rely on artificial neural networks (ANNs) that try to simulate how the brain works to learn.
ANN can be trained to recognize patterns of information such as voice, text data, and visual images, and is the basis of many recent developments in AI.
Traditional AI “teaches” algorithms on a particular subject by using input to provide a large amount of information.
AI systems rely on artificial neural networks (ANNs) that try to simulate how the brain works to learn. ANN can be trained to recognize patterns of information such as audio, text data, and visual images.
Practical applications include Google’s language translation service, Facebook’s facial recognition software, and Snapchat’s live image modification filter.
The process of entering this data can be very time consuming and is limited to one type of knowledge.
A new type of ANN called Adversarial Neural Networks allows the wisdom of two AI bots to match each other and learn from each other.
This approach is designed not only to improve the output produced by the AI system, but also to speed up the learning process.
Scientists read bird’ brain signals to predict what they’ll sing next Source link Scientists read bird’ brain signals to predict what they’ll sing next
The post Scientists read bird’ brain signals to predict what they’ll sing next appeared first on California News Times.