Voice Activated Controls
bradpitcher at gmail.com
Tue Feb 27 01:43:32 CET 2007
But who's going to pay for the infrastructure required to support these
On 2/26/07, Dean Collins <Dean at cognation.net> wrote:
> The answer is no, the neo processing power is too limited.
> You need to start thinking bigger guys.
> Why does the processing need to occur on the handset itself? What about
> building a 'cloud' application where the application and processing occurs
> with only a sip or gsm connection to the handset?
> Check out http://www.voip-info.org/wiki/view/Tellme for an idea on what I
> am describing.
> Dean Collins
> Cognation Pty Ltd
> dean at cognation.net
> +1-212-203-4357 Ph
> +1-917-207-3420 Mb
> +61-2-9016-5642 (Sydney in-dial).
> *From:* community-bounces at lists.openmoko.org [mailto:
> community-bounces at lists.openmoko.org] *On Behalf Of *Brad Pitcher
> *Sent:* Monday, 26 February 2007 7:07 PM
> *To:* community at lists.openmoko.org
> *Subject:* Re: Voice Activated Controls
> Perhaps it could be used as a form of input for typing, instead of the
> virtual keyboard. I would love to use it for sending texts or emails.
> On 2/26/07, *Knight Walker* <moko at kobran.org> wrote:
> On Mon, Feb 26, 2007 at 05:08:18PM -0600, Jonathon Suggs wrote:
> > Does anyone know of any software for natural language processing that
> > could be ported to OM/Neo? I really like some of the software that is
> > available for the PocketPC (MS Voice Commander and Fonix). They both
> > run and work well on a resource limited platform as well, so it *can* be
> > done, but both are closed.
> > Here are a couple of OS engines:
> > http://www.speech.cs.cmu.edu/pocketsphinx/
> > http://julius.sourceforge.jp/en_index.php?q=en/index.html
> > http://xvoice.sourceforge.net/
> > So I guess, is there already any voice control software planned/worked
> > on for use in OpenMoko? If not, I'll help out, but can we get a project
> > up and running?
> Depending on what you're looking for, it may be easier than that. My last
> three phones (All Nokia) have a speech command recognition which seems
> more like limited wave-form matching. It requires that I record the
> command ("Call Brian", "Start music player", "Begin recording") then when
> I want to use that command, I hold down a button on the phone (Or on the
> headset) until the phone makes a certain tone, and speak the command.
> The phone then "searches" for a few seconds, finds the match, and
> activates the function. It's VERY useful for when I'm driving and want
> to call someone. I just hold down a button on my headset and give it the
> command, all while keeping my eyes on the road.
> My only major complaint is the (very) limited number of commands you can
> record. I find I have to think really hard about what I want a voice
> command for because making a new one requires that I remove an existing
> It wouldn't give you continuous speech-to-text capability, but I'm not
> sure I want that. People think I'm crazy enough as it is.
> OpenMoko community mailing list
> community at lists.openmoko.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the community