[LAD] entailments of probabilistic processing chip for DSP and audio processing?

Charles Henry czhenry at gmail.com
Tue Aug 17 23:38:27 UTC 2010


hmmm... interesting

The discussion is mostly about improvements to existing algorithms
such as error checking and correction and encryption.  So, it leads me
to ask if there are any such problems in the DSP field that could be
addressed with it?  Please, please don't let it be DRM :P

Seems like markov processes and Bayesian modelling problems could be
implemented (although I'd really like to see the math btw input/output
on the gates before I conjecture--I'll do so anyway.)  Speech
analysis/recognition could be made less costly.  Maybe there could
even be a small speech co-processor for mobile devices.

or how about source separation and the ever elusive automatic
transcription problem?

Chuck

On Tue, Aug 17, 2010 at 5:48 PM, Niels Mayer <nielsmayer at gmail.com> wrote:
> http://www.technologyreview.com/read_article.aspx?id=26055&a=f
>
> A computer chip that performs calculations using probabilities,
> instead of binary logic, could accelerate everything from online
> banking systems to the flash memory in smart phones and other gadgets.
> [...]
> The electrical signals inside Lyric's chips represent probabilities,
> instead of 1s and 0s. While the transistors of conventional chips are
> arranged into components called digital NAND gates, which can be used
> to implement all possible digital logic functions, those in a
> probability processor make building blocks known as Bayesian NAND
> gates. Bayesian probability is a field of mathematics named after the
> eighteenth century English statistician Thomas Bayes, who developed
> the early ideas on which it is based.
> [...]
> Lyric has been working on its technology in stealth mode since 2006,
> partly with funding from the U.S. Defense Advanced Research Projects
> Agency. DARPA is interested in potential defense applications that
> would involve working with information that isn't clear cut--for
> example, radio signals distorted accidentally or otherwise, and
> machine vision systems that try to recognize actions or objects in
> images. "They're interested in some James Bond-type applications,"
> says Vigoda.
>
> Niels
> http://nielsmayer.com
> _______________________________________________
> Linux-audio-dev mailing list
> Linux-audio-dev at lists.linuxaudio.org
> http://lists.linuxaudio.org/listinfo/linux-audio-dev
>



More information about the Linux-audio-dev mailing list