hi all,
helping a friend with an installation ,he wants to integrate voice recognition
where the mic connected to the computer will display what a person(s) is saying
on a t.v. or projector
another friend told me there are max objects (externals) that do this
but i want to do this with VVVV
any links or help would be great
mh, vvvv has the FFT node to analyze sound, you can display this data like you want, but doing a recognition would be hard, you have to patch it … another idea is that you do the recognition with a simple patch in max (or look if its also for pd available) and send the data to vvvv with netsend.
you can see in my music3d patch a simple FFT visualization.
In a much older version of vvvv there was an implementation of the viaVoice engine. But I’d say that’s not entirely easy to reproduce.
Have you tried to find out how a speech recognition engine (like “naturally speaking”, “viaVoice” etc) talks to an application? chances are that it actually simulates keytrokes. You could intercept those easily.
the other way was simpler: Voice _output_can be done via a Javascript that you send to a renderer (HTML) using the MS Agent API (google that)… perhaps that’s even one way to go for recognition, I believe MS Agent does that, too. But that’s just a guess.
(clik clik) ah, here it is:
Driven by a powerful yet simple programming model, these onscreen personalities can gesture, move, speak (via a text-to-speech engine or recorded audio) and even accept spoken voice commands. http://www.microsoft.com/msagent/prodinfo/default.asp
erm, and you can disable the “onscreen personalities”, too.
thanks for the quick response
via vocie was one of the programs iwas looking
at,i dont remember a via vocie node (only been useing VVVV since 33beta 4)
but sounds interesting,an internal solution and all.
i think i will look into pd for getting it in VVVV
nice ms agent link.
thanks to both of you
will post the patch once i get it working