Extraordinary Ability of Blind People to Hear Ultrafast Speech
New research presented at the Society for Neuroscience meeting in San Diego shows that blind people can understand speech at ultrafast rates, well beyond what a sighted person can comprehend. Using brain imaging, the researchers discovered how they were able to do this. The parts of the brain that process hearing get re-wired to the part of the cerebral cortex that normally handles vision. This is explained in my post on the Scientific American website, but Scientific American was not able to include the audio clip of what such high-speed speech sounds like. Have a listen here, and read the story at Scientific American News on-line: “Why can some blind people process speech far faster than sighted persons: www.scientificamerican.com/article.cfm?id=why-can-some-blind-people -process
Blind people in the study could hear speech as fast as 25 syllables a second!
For those of us who are sighted, here is a transcript of the sound clips:
“Blackwater, now called Xe Services, was once the United States’ go-to contractor in Iraq and Afghanistan. It has been under intense pressure since 2007, when Blackwater guards were accused of killing 17 civilians in Nisour Square in Baghdad. The company, its executives and personnel have faced civil lawsuits, criminal charges and congressional investigations surrounding accusations of murder and bribery. In April, federal prosecutors announced weapons charges against five former senior Blackwater executives, including its former president.”
My thanks to Dr. Ingo Hertrich, University of Tuebingen for generating these sound clips in English for readers to hear.
Thanks for the new English speech samples, Douglas! Some blind people I know confirmed that the 24 syllables per second sample is at their upper limit. I am sighted and max out already at the 14 sps sample, but perhaps I can blame it on not being a native English speaker – holding hopes that I would do better with Dutch. 🙂
Re your comments to your Scientific American article,
> But does such extraordinary ability come at a cost? This work is also interesting
> with respect to synesthesia. Do blind people see speech since auditory
> pathways get rewired to visual cortex?
Don’t know yet about cost or penalties, but some partial and preliminary answers to your questions can be found in
Fascinating. Read your piece in the Huffpost – also right on target and very good read when supported by scientific pursuit. I see lots of behavioral toxicity and many times the owner of the problem is not at all aware of how deep rude behaviours serve to telegraph intent.
My interest in sound pattern recognition is similar, characterizing musical phrases as methods of choice in the possession of the musician ( scoring film and tv music properly is best done when based on immediate cognitive responses )
Buying the book now. I hope you choose to do more in auditory response, since that retained memory is rather deep and sociocultural as well…
Do you think it is possible for a sighted person to increase the number of syllables per second that we can process? For example, regularly listening to 10 sps speech over a course of several weeks, then 11 sps, then 12 and so on?
Hi Tim, Sorry for the late response due to a glitch. Of course you can improve with practice, but no sighted person is known who can attain the listening speeds of these individuals who are using their visual cortex for auditory processing.
Is it really that this man can pronounce 20 syllables per second? I think many auctioneers surpassed him in pronouncing most syllables. Try searching ‘Kevin McGlothlen’ in Youtube, he is such an unbelievable auctioneer.
http://www.youtube.com/watch?v=wlPAuYItZm0 (New Guinness record/ 20 syllables per sec. 852 in 42 secs)
Hi, I’m 43, and totally blind since birth. A blind friend from Canada forwarded me the original article, and I got here from the comments. The 14sps sample is near the top of my comfortable range, I have friends who run their speech at 21 to 24 all the time, but I only get a word or phrase here and there at those speeds. Though I can recognize phrases I hear from my screen reader on a regular basis at those speeds, for instance, names of applications such as “Mozilla Firefox”, information about things on a webpage like “checkbox unchecked”, or “link”. The speed at which I can read a given thing depends on what kind of thing it is. If it’s routine stuff about what my computer is doing, or I’m playing a simple card game, or some other activity where there’s a lot of repetition of the same information, I can comprehend that stuff at much higher speeds than, say, reading a simple email from a friend. If I’m reading something technical, like a science article, I have to slow down even more.
I imagine sighted people have a similar experience when they read. I certainly do when I read braille, though I can’t read braille at anything near the speed at which I can understand speech.
I also speak Spanish and Portuguese as second and 3rd languages, and a bit of Esperanto. I can’t read as fast in those languages either, but what I said above applies to them, Proportionately at slower speeds of course. I’m also not as fast with other synthesizers I’m not used to using. I typically use eloquence, the same synthe as in the clips. If I listen to DecTalk or ESpeak, for example, or a synth created using a human voice, I can’t go as fast. But if I use it long enough, certainly my speed would improve. I, and I know I’m not the only one, find formant synths easier to comprehend when speeded up than concatenated ones.
On another note, my digital talking book player allows me to speed up the voices of the human narrators without changing the pitch, and my speed with that depends on the narrator. For some reason I seem to be able to listen to men faster than women, even though I’m not changing the pitch at all. Back in the days before digital processing, we could speed up our cassette players, but the pitch went up too. Imagine chipmunks on redbull. Often I could listen at double speed, or very near double speed for a deep voiced narrator. Any faster, and it’s like trying to understand deep sea divers on heleox.
Thank you for sharing your experience, Shawn. I have recently met a remarkable person with congenital blindness who can do this and much more. I love how this shows the untapped potential of the human brain.
A pleasure. I thought someone might find my observations useful. Another thing I can do, understand English text spoken by a synth using rules for a different language, especially, but not exclusively, Spanish and Portuguese. I can also read Spanish and Portuguese with an English speaking synth, especially Eloquence, (others I don’t use every day, not so easily), and strangely enough, sometimes I can do it faster that way for simple texts, but I try not to cheat like that too much, as it encourages bad habits lol. Most bi/multilingual blind people I know can do this. In fact I have to cajole friends who are learning English into reading it with the propper synth, because they often pick up bad habits from reading English with a synth designed for their native language. Most Pakistani blind that I know read and right Urdu using the English synth, because until ESpeak, there wasn’t a good affordable one that spoke Urdu.
It is amazing what the human brain can do. I remember a brain teaser email that was going around a few years ago about how you can understand whole sentences where all the words are scrambled, as long as the first and last letters are where they belong. For emaxlpe, it deson’t mttaer in waht oredr the ltteers in a wrod aepapr, the olny iprmoatnt tihng is taht the frist and lsat ltteer are in the rghit pcale. The rset can be a toatl mses and you can sitll raed it wouthit pobelrm.
It doesn’t work as well if you read it with a synthesized voice, but for many words (particularly ones that don’t have silent combinations like gh) it does. But if I read letter by letter, no problem at all.