Return to the archive index

Re: CMU Sphinx Speech Recognition

From: Vito <>
Date: Fri, 04 Feb 2000 08:24:38 -0500

Scott Pircher wrote:
> 
> Personally, one of the eventual goals of my pipedream wearable is to
> record all my conversations, both what I'm saying, and what everyone
> around me is saying. I intend to implemetn something akin to an IRC
> Infobot that'll listen to the conversations (alot easier with speaker
> independent voice recog than typing them in myself) and store
> hopefully relevant information for later retrieval whenever I ask it
> questions.

Good, except why have the wearable do it?  Why not have the system
simply record what was said and feed it back to a more capable desktop
or network system to post-process the data?

Personally, I always figured my pipedream wearable ("Jane," from the
name of the computer-based entity who communicated to Ender Wiggin
through an earring-like stub, in the Ender's Quartet series by Orson
Scott Card) to be very low-tech.  The core components would be a
microphone and earphone, and a wireless data link, and it would feed my
requests to a remote system which would process my voice and aurally
return the data to me.  None of that was ever actually done on the
wearable itself.  The rest of the wearable would be optional add-ons
like storage (for on-person libraries of information like manuals or
maps), GPS, additional data links for greater speed, video, etc.

--Vito

--
Subcription/unsubscription/info requests: send e-mail with subject of
"subscribe", "unsubscribe", or "info" to 
Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org

+Previous Message in Thread | Next Message in Thread

From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty

Archive created with babymail