Voice-Recognition & Linux. Still a fantasy?? - Powerpc

This is a discussion on Voice-Recognition & Linux. Still a fantasy?? - Powerpc ; Voice-Recognition & Linux. Still a fantasy?? Hi Linux people Pardon me for multi-posting this to various groups but I need to collect feedback from as wide a spread as possible. My university is investigating Voice Recognition Software, for use by ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: Voice-Recognition & Linux. Still a fantasy??

  1. Voice-Recognition & Linux. Still a fantasy??

    Voice-Recognition & Linux. Still a fantasy??

    Hi Linux people

    Pardon me for multi-posting this to various groups but I need to
    collect feedback from as wide a spread as possible.

    My university is investigating Voice Recognition Software, for use by
    disabled students. Considering that this particular application of the
    technology would be highly demanding (wide lexicon, varied disciplines
    and terminology in several languages) am I right in assuming that it's
    going to stay under the category of A GREAT IDEA THAT WON'T WORK??

    I'd appreciate any comments, positive and negative. Thanks...

    Fintan

  2. Re: Voice-Recognition & Linux. Still a fantasy??

    > My university is investigating Voice Recognition Software, for use by
    > disabled students.


    Speech recognition is one of those areas that doesn't lend itself to Open
    Source development (due to the highly theoretical and technical nature of
    the alogrithms involved, and since there is little immediate benefit from
    voice recognition until such time as it is nearly perfect).

    Likewise, due to limited Linux desktop penetration, the big players in
    consumer speech recognition have not made major inroads into the Linux
    platform (vis. IBM's withdrawl of support for ViaVoice).

    The real action is in APIs, generally targeted at the telephony market. For
    example:
    * Babel Technologies' BaBear
    http://www.babeltech.be/Products.php?s=77&m=75&f=83
    * ScanSoft's OpenSpeech Recognizer
    http://www.scansoft.com/openspeech/recognizer/
    * Nuance's Nuance 8.0
    http://www.nuance.com/prodserv/prodnuance.html

    Integrating a voice recognition API into a few specific applications is not
    overly difficult, and certainly that is the kind of task that a university
    would find the funding for. Having the entire OS voice-steerable is
    somewhat more of a challenge, one not currently addressed within Linux.

    Let us know how you go.

    Paul

    --
    P-a-u-l- -J-a-n-k-o-w-s-k-i pmj@users.sf.net

    "A programmer," he said with obvious amazement, is the sort of person "who
    drinks Coke in the morning." -- Boston Globe article on The High Tech Set


  3. Re: Voice-Recognition & Linux. Still a fantasy??

    In article <3f680dc6@news.alphalink.com.au>, Paul Jankowski
    wrote:

    > > My university is investigating Voice Recognition Software, for use by
    > > disabled students.

    >
    > Speech recognition is one of those areas that doesn't lend itself to Open
    > Source development (due to the highly theoretical and technical nature of
    > the alogrithms involved, and since there is little immediate benefit from
    > voice recognition until such time as it is nearly perfect).
    >
    > Likewise, due to limited Linux desktop penetration, the big players in
    > consumer speech recognition have not made major inroads into the Linux
    > platform (vis. IBM's withdrawl of support for ViaVoice).
    >
    > The real action is in APIs, generally targeted at the telephony market. For
    > example:
    > * Babel Technologies' BaBear
    > http://www.babeltech.be/Products.php?s=77&m=75&f=83
    > * ScanSoft's OpenSpeech Recognizer
    > http://www.scansoft.com/openspeech/recognizer/
    > * Nuance's Nuance 8.0
    > http://www.nuance.com/prodserv/prodnuance.html
    >
    > Integrating a voice recognition API into a few specific applications is not
    > overly difficult, and certainly that is the kind of task that a university
    > would find the funding for. Having the entire OS voice-steerable is
    > somewhat more of a challenge, one not currently addressed within Linux.
    >
    > Let us know how you go.
    >
    > Paul
    >

    Thanks Paul, for that informative advice. I'll certainly keep you
    posted. My Uni is one of the worst hit by the recent Govt atrocities,
    so cash is certainly short. My own view (prior to whatever I may learn
    in this investigation) is that if they are intending to support
    research in this area (and call it commercial or not, that is the
    reality at this stage, not paying for a slick finished product) then it
    should go to support the Linux effort.

    You've certainly endorsed my suspicions that the suggestion that the
    entire OS could be successfully controlled in this way (which is the
    suggestion) is little more than A Nice Idea.

    Fintan Mulcahy

+ Reply to Thread