Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
Input Devices Intel Open Source

Stephen Hawking's New Speech System Is Free and Open-source 56

An anonymous reader writes: Stephen Hawking and Intel have worked together for the past several years to build a new communication system for those suffering from diseases that severely impair motor function. The system is called ACAT (Assistive Context Aware Toolkit), and it will be free and open source. Hawking's previous system had been in use for over 20 years, so the technological upgrade is significant. His typing rate alone has doubled, and common tasks are up to 10 times faster. ACAT uses technology from SwiftKey, a cell phone keyboard enhancement.

"Over three million people around the world are affected by motor neuron disease and quadriplegia and because the system created for Hawking is based on open-source software, it could potentially be adapted to suit many of them. Different functions can be enabled by touch, eye blinks, eyebrow movements or other user inputs for communication. Hawking and Intel hope that because the system is open and free it will be adopted by researchers who will want to use it to develop new solutions for those with disabilities."
This discussion has been archived. No new comments can be posted.

Stephen Hawking's New Speech System Is Free and Open-source

Comments Filter:
  • The software being open source is definitely a good step, especially since it means there can now be open research on improving the system, whereas much of the previous research was done on proprietary systems.

    Does anyone know much about the hardware side, though? E.g. are we talking $10k of equipment, $100k of equipment, or some entirely custom special-ordered system?

  • by MouseR ( 3264 ) on Tuesday December 02, 2014 @04:31PM (#48510263) Homepage

    Being open-sourced it'll get forked and in no time Hawking will be able to communicate with the reptilians.

  • by Anonymous Coward

    Epic prank calls in 3... 2.... 1...

    [Hawking voice] "Hello this is Ste-phen Haw-king..."

    • by ArcadeMan ( 2766669 ) on Tuesday December 02, 2014 @05:53PM (#48510847)

      Moe: [answering the phone] Moe's Tavern.
      [Hawking voice]: Hello. Is HAL there?
      Moe: HAL?
      [Hawking voice]: Yes, HAL. Last name: 9000.
      Moe: Let me check... [calls] Phone call for HAL. HAL 9000. Is there an HAL 9000 here?
      [bar patrons laugh]
      Moe: Wait a minute. [to phone] Listen, you little wheelchair jackass, if I ever find out who you are, I'll kill you!
      [Hawking voice]: Ha. Ha. Ha.
      Homer: I hope you do find that punk someday, Moe.

  • download link? (Score:5, Insightful)

    by atheos ( 192468 ) on Tuesday December 02, 2014 @04:37PM (#48510313) Homepage
    and we can get it where?
  • MC Hawking? (Score:5, Funny)

    by xaotikdesigns ( 2662531 ) on Tuesday December 02, 2014 @04:46PM (#48510385) Homepage Journal
    Maybe this means we can get another MC Hawking album
  • did anyone find themselves reading the summary in the Stephen Hawking machine voice? Every time his name comes up, I hear the voice.
  • by Immerman ( 2627577 ) on Tuesday December 02, 2014 @04:50PM (#48510407)

    This new software is great and all, but I've got to ask - with all the advances in neural interfaces, why haven't I heard of an alternative communication system harnessing them? I mean we've got monkeys that have rapidly learned to control a robotic arm using only signals from a tiny cluster electrodes in their brain, essentially granting them a whole new virtual appendage. It seem like it shouldn't be terribly difficult to do the same thing for someone like Hawking - stick some electrodes in his brain and use the signals to control a cursor or six. It may take him a bit of biofeedback practice to get conscious control over them, but then he'd have a fast and versatile N-axis input device to drive whatever systems he's using.

    Granted, at this point his motor cortex has probably largely atrophied what with the signal lines having long gone dead, and I could understand not wanting to tamper with the more cognitive portions of his brain, but surely there's some spot that would still be serviceable. As I understand it it could probably even be some completely random place, practice and neuroplasticity will see to converting the cells being monitored into output nodes.

    • by Guspaz ( 556486 ) on Tuesday December 02, 2014 @04:57PM (#48510457)

      RTFA (second link): they've tried it, and couldn't get a strong enough signal from Hawking.

      • Now why would I do that? I read the articles that promise to be interesting, and offer my insights to those who ask. And I ask my own questions in turn on those articles that I do not understand or can't be bothered to read. Everyone wins. In this case nothing in the summary suggested that the article would even address my question, and I'm uninterested in the details of a low-bandwidth communication software, even when it's open source.

        Thanks for the answer though, despite the rude delivery. Skimming the

    • by JanneM ( 7445 )

      "we've got monkeys that have rapidly learned to control a robotic arm using only signals from a tiny cluster electrodes in their brain,"

      "rapidly" and "control" are very much relative terms in this case. And note the "in their brain" - you need to implant an electrode array to get good, reliable signals. With monkeys you can do it to half a dozen animals and hope than one or two get a fully working implant. And the array has to be working for a few months or so. With a human patient you need to get it right

      • As I understand it the technology has been refined considerably, and is being used routinely and to lasting effect with experimental prosthetics likely to hit the market within a few years.

        It's difficult if you're trying to monitor (or stimulate) specific cells, but the typical probe is a long thin rod composed of dozens if not hundreds of electrodes of different lengths. You can then close up your patient and simply monitor/stimulate the electrodes individually to find the ones best suited to your intende

  • by JustNiz ( 692889 ) on Tuesday December 02, 2014 @05:01PM (#48510485)

    >>> Stephen Hawking's New Speech System Is Free and Open-source

    Maybe my boss will take my ideas more seriously when I sound like Stephen Hawking.

  • Where's the link to the source and the schematics instead of the PR pieces ?
  • Coincidentally, the following was at the top of this story when I loaded it:

    Slashdot stories can be listened to in audio form via an RSS feed, as read by our own robotic overlord.

    I gave it a go, and the results are not exactly brilliant. For example, 'sujan.sun writes "Like clockwork, the first...' was read out as 'sujan dot sun writes like clockwork. The first...'

    Even my Kindle Keyboard, which has pretty decent TTS otherwise, has terrible timing when it comes to punctuation.

    http://www.ivona.com/ [ivona.com] do the best I've heard so far. British Amy can give me turn-by-turn navigation any day.

  • by account_deleted ( 4530225 ) on Tuesday December 02, 2014 @05:31PM (#48510705)
    Comment removed based on user account deletion
    • by pavon ( 30274 ) on Tuesday December 02, 2014 @06:07PM (#48510939)

      From the second link:

      Wood showed WIRED a little grey box, which contained the only copy of Hawking's speech synthesiser. It's a CallText 5010, a model given to Hawking in 1988 when he visited the company that manufactured it, Speech Plus. The card inside the synthesiser contains a processor that turns text into speech, a device that was also used for automated telephone answering systems in the 80s.

      "I'm trying to make a software version of Stephen's voice so that we don't have to rely on these old hardware cards," says Wood. ...

      Hawking is very attached to his voice: in 1988, when Speech Plus gave him the new synthesiser, the voice was different so he asked them to replace it with the original. His voice had been created in the early 80s by MIT engineer Dennis Klatt, a pioneer of text-to-speech algorithms. He invented the DECtalk, one of the first devices to translate text into speech. He initially made three voices, from recordings of his wife, daughter and himself. The female's voice was called "Beautiful Betty", the child's "Kit the Kid", and the male voice, based on his own, "Perfect Paul". "Perfect Paul" is Hawking's voice.

  • Stephen Hawking: I call it a Hawking-Voice.

  • In the US the software may end up being free but the hardware to run it will be $10K+ because this is part of a medical device and we have bureaucratic ticks to check off for "safety".

"The vast majority of successful major crimes against property are perpetrated by individuals abusing positions of trust." -- Lawrence Dalzell

Working...