[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: gperiodic



Kyle wrote:

> The way it's implemented right now, it's all optional.  If anyone wants
> to contribute sound files, I'll put them up on the gperiodic website,
> and offer them up as an option -- as a seporate package, with it's own
> install routines.

The problem is having the hardware to do it. Someone once showed me a device
that one can carry around and speak into. When back at the computer, one can
connect it to a serial port and download the .wav files. I'm assuming one
could create multiple .wav files. Again I'm assuming we want a .wav file for
each element. I will see if I can find this device. However it was over a year
ago. I guess one could plug a microphone into a sound card but one would also
need software to go with it. It would be nice to at least try gperiodic out
with sound.

Roger wrote:

> I wonder about comparing audio clips, so the user can mention an
> element and it flashes, and then zeroes in on the one he said.
> But that's for a different month, I should think. :)

This gets into speech recognition which we need to look into. I guess with
only a hundred or so words, this might not be to hard. Unfortunately most
speech recognition work is proprietary. However it is a coming thing. Again it
would be nice to be able to try it out.

Bob