[MUD-Dev] Shift in time

[Thomas Clive Richards]

On Sunday October 3 2004 11:16 pm, Arnau Rossell=F3 Castell=F3 wrote:

> Today is a good day to de-lurk.

Aye ;)

> Would such a thing be feasible?  How much bandwidth would the
> phoneme stream need(well i could time myself saying something and
> then count... but maybe someone has more elaborate aproximations)?
> How could such a system deal with ambient noise, or tapping the
> mic?  Is there any king of public project(i know about festival
> only), creating these kind of programs? papers maybe?

> I've searched on my(partial) log of mud-dev, but i've only found
> talk about the design side of this.

I've been considering a very similar idea myself.

Unfortunately, voice recognition in *any* form is *VERY* hard to
accomplish. The method you describe is actually very similar to the
way leading voice recognition apps do it.

The trouble is creating an algorithm that translates from human
voice to phoneme strings. Even the best voice recognition packages
need extensive training. If you happen to have a thick Slavic,
Indian, or Yorkshire accent (for example), it can be very hard to
get decent accuracy.

There was an articale on gamasutra ages ago, but I can't find it
anymore... I guess they remove old articles after a few years.

HTH!

--

Thomi Richards,
thomi at once.net.nz
_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev