I found that while the technology is speaker independent, it seems to take a few tries before it matches what you are saying with the list of phrases.  What happens with me is I end up repeating myself two to three times very distinctly per phrase and then recognition occurs at a higher frequency without the repetition or deliberate pronunciation.  The are more examples on the developer CD that are better suited for recognition but they are written in Cocoa, which I don't understand.

I think I asked about increasing recognition on Apples speech recognition list but didn't get a response.  That list hardly gets used.  I subscribed to it months ago.

I have a southern draw.  I let my wife (Alicia) test the program out and she had little (notice I didn't say no problem) problem with the program recognizing her speech and she doesn't have southern accent.

<<
I have my microphone selected as an input device (and have played with 
the Input Volume), and I presume the phrases/words that can be 
recognized are the ones in the the DATA statments:
>>

Yes, the phrases in the data statement are the ones recognize.  I thought data statment
made it easy to change and experiment with.

I wish I had more time.  Family, work and school are taking a *major* toll on me.
I originally had a specific plan for using speech recognition but I've tabled pretty much everything except for programming that's in line with my education track. From time to time I get a treat and get to work on something fun.  I havent been able to revisit that code to see what I could have done better or make it more user friendly.


W.


-----Original Message-----
From: Darron Forehand [mailto:glo@...]
Sent: Tuesday, November 09, 2004 5:26 PM
To: futurebasic@...
Subject: Re: [FB] Speech Recognition


Waverly,

This really looks like a cool and fun example, but I've tried more than 
a few times to get it to respond without any luck.

I have my microphone selected as an input device (and have played with 
the Input Volume), and I presume the phrases/words that can be 
recognized are the ones in the the DATA statments:

> data "Hi"
> data "Goodbye"
> data "What time is it"
> data "What day is it"
> data "testing"
> data "bubble"
> data "THIS IS ALICIA"
> data "Are we there yet?"
> data ""

Any tips or advice on how to make it work properly?  I'm hoping that my 
Texas drawl doesn't cause me to be descriminated against lol.

Darron



Edwards, Waverly wrote:

>I translated the speech recognition example from Apple's site last month.
>I added a little extra just to make the recognition prove itself.
>
>It works perfectly, however if someone would be kind enough to point out
>to me how I could have fixed this line of code better I would appreciate it.
>
>if (not rStatus ) *model = newModel
>
>The problem is that we are returning newModel by reference.  Because
>I couldn't get it right I ended up adding a global and turned it into this
>
>if (not rStatus) then gModel = newModel
>
>Enjoy,
>
>
>W.
>
>
>
>ftp://ftp.futurebasic.org/FutureBasic/dropbox/Speech%20Recognition.sit
>
>
>
>
>--
>To unsubscribe, send ANY message to: futurebasic-unsubscribe@...
>  
>

--