Re: PC Speech Recognition

[Chris Hunter <cjhunter@xxxxxxxxxxxx>]

Phil Harris wrote:

>>>Looking into it how deeply?
>>>
>>>
>>>
>>enough to work-out how far we should go & what appraoch we
>>should take (what to rely on it for)
>>
>>
>
>But you also seem to have been restricting yourself to the budget end
of
>voice recognition solutions ...
>
of course ...

>>we don't want to subject guests to a training programme whe
>>the arrive !
>>
>>
>
>You're going to have to train them when they arrive anyway as they'll
be
>looking for light switches and remotes...
>
>
yes, but it should be intuitive !

>>>Perhaps find out the issues they have rather than presuming
>>>
>>>
>>what they might
>>
>>
>>>be?
>>>
>>>
>>>
>>thanks !
>>
>>
>
>Well, doesn't it seem sensible to find out what the issues are with it
and
>then you can find out whether they can be addressed or whether you need
to
>look at other solutions...
>
I meant thanks for assuming I wasn't doing that !!

>>>The are some excellent audio output cards available for PC
>>>
>>>
>>if you need high
>>
>>
>>>quality output and in the case of input (which is more
>>>
>>>
>>relevent here) then
>>
>>
>>>there are all manner of quality solutions at prices ranging
>>>
>>>
>>from the merely
>>
>>
>>>eyebrow-lifting to the criminally insane. If you wish you
>>>
>>>
>>can even do all
>>
>>
>>>the analogue to digital and digital to analogue conversion
>>>
>>>
>>away from the PC
>>
>>
>>>and simply feed bitsreams of audio in or out so I don't see
>>>
>>>
>>what is to be
>>
>>
>>>desired - perhaps you simply mean at the budget you are
>>>
>>>
>>willing to spend?
>>
>>
>>of course there are, but that's not what people are generally
>>using ...
>>
>>
>
>Here I think you are barking up the wrong tree ... Standard PC audio
>hardware (the analogue to digital conversion process) is more than good
>enough quality to handle the fairly meagre demands placed upon it by
voice
>recognition. It's the analogue pre-processing (gates, limiters, mixers)
and
>the digital post processing (software analysis of captured data) that
is the
>"problem"
>
maybe ... but I suspect programming techniques are also part of the
problem ... and I'm sure  stereo should be there, too, to help
distinguish sounds ...

>>location is part of it, but distinguishing is more the point here
...
>>
>>
>
>Location is pretty easy without requiring "stereo" - at its
simplest it can
>be determined by which room mic is picking up the voice.
>
>I don't understand how feeding stereo mics into something like ViaVoice
will
>help ViaVoice distinguish what you are saying.
>
useful in distinguishing sounds ...

>Phil
>
still, there are some applications of VR around that really do work (eg:
in Typhoon fighters) ... so I'm sure we'll find a way through this !


[Non-text portions of this message have been removed]