[MV] Re: Questions on customising VV

[Camner]

To echo and emphasize:

Good speech recognition and accuracy REQUIRES one to use "audio setup" on a
regular basis (ideally each time one starts up VV).  I don't always do that.
Accuracy always improves when I do so.

On 5/29/03 12:26 PM, "T. Patrick Henebry" <tphenebry at comcast.net> wrote:

>> If I go into UserSettings, I see a pop-up w/ "...first voice model"
>> selected, and options to add, modify, or rename it ('delete' is
>> inactive).  I want to COPY that model - make a duplicate - and treat
>> them as separate models, even though the hold the same data, much as
>> one might duplicate a file & handle the copy as if it was a separate
>> document.  I want to label one of them LOUD, & set the mic up for
>> loud conditions...and label the other QUIET & set the mic up for
>> quiet conditions.  I would then select one or the other, and work
>> with each independently from the other.
>> 
>> To repeat:  has anyone tried anything like this - and, if so, how
>> did it work out for you?
> 
> Charlie,
> 
> No speech recognition program would allow you to do that. In order to
> maintain the best possible accuracy they all are designed so that
> microphone levels can be adjusted based on the current conditions at
> the time of running the microphone setup so the clearest possible
> signal is available for processing by the recognition engine. This
> way the best possible signal is guaranteed no matter what voice model
> is being used.