Synthetic Speech Made from Brain Recordings

UC San Francisco neuroscientists can generate natural-sounding synthetic speech by using brain activity to control a virtual vocal tract – an anatomically detailed computer simulation including the lips, jaw, tongue and larynx. The study was conducted in research participants with intact speech, but the technology could one day restore the voices of people who have lost the ability to speak due to paralysis and other forms of neurological damage.

Stroke, traumatic brain injury, and neurodegenerative diseases such as Parkinson’s disease, multiple sclerosis and amyotrophic lateral sclerosis (ALS, or Lou Gehrig’s disease) often result in an irreversible loss of the ability to speak. Some people with severe speech disabilities learn to spell out their thoughts letter-by-letter using assistive devices that track very small eye or facial muscle movements. However, producing text or synthesized speech with such devices is laborious, error-prone, and painfully slow, typically permitting a maximum of 10 words per minute, compared to the 100 to 150 words per minute of natural speech.

It is possible to create a synthesized version of a person’s voice that can be controlled by the activity of their brain’s speech centers. In the future, this approach could not only restore fluent communication to individuals with severe speech disability, the authors say, but could also reproduce some of the musicality of the human voice that conveys the speaker’s emotions and personality.

Nature – Speech synthesis from neural decoding of spoken sentences

9 thoughts on “Synthetic Speech Made from Brain Recordings”

  1. That is a lot of effort. You have to implant these things. And you probably can’t just put it in and turn it on. You probably have to cooperate to have it map everything correctly.

  2. Standard procedure when entering the USA is to make a signed statement that you are not there to commit terrorist actions or stuff. Now take that exact question and add a magic sorting hat while you answer.

    And they already do stuff like fingerprints and body scans, so it isn’t as if they would hesitate for privacy reasons.

    Step 2 is to include a similar statement about safe driving whenever you renew your driving licence. It’s to protect the Children.

    Step 3? Well don’t you guys already have to swear an oath of allegiance every morning at school? Or is that a fictional exaggeration?

  3. Hint: a lot of them casually commit crimes everyday.

    Such as:

    Attempt to evade or defeat paying taxes: Upon conviction, the taxpayer is guilty of a felony and is subject to other penalties allowed by law, in addition to (1) imprisonment for no more than 5 years, (2) a fine of not more than $250,000 for individuals or $500,000 for corporations, or (3) both penalties, plus the cost of prosecution (26 USC 7201).

    Fraud and false statements: Upon conviction, the taxpayer is guilty of a felony and is subject to (1) imprisonment for no more than 3 years, (2) a fine of not more than $250,000 for individuals or $500,000 for corporations, or (3) both penalties, plus the cost of prosecution (26 USC 7206(1)).

  4. It’s a step towards telepathy.

    I see they can also make you hear sounds by hitting you with a laser these days. They are also making rapid strides in bionic eyes.

    So bionic eyes with lasers and a thought to speech interface? Pretty cool possibilities.

  5. Food for worry ? Maybe for spies, if it can’t be trained against, but I don’t see how this should worry the general population.

  6. Hmmm ….! I wonder further development could be used to record your inner dialogue. For good or ill it would be a powerful interrogation apparatus. Ask a question and then asks if the response was the truth. The inner dialogue may “spill the beans” in an almost involuntary way! Food for Worry!

Comments are closed.