Zabaware Support Forums

Haptek LipSync Stuff

Started by Art, June 07, 2005, 03:31:15 PM

Previous topic - Next topic

Art

OK...previously we discussed the Haptek lipsync, OGG files, etc., BUT, did you know that you can enter ANY text you want the Haptek character to say?

Simply create a text file of your choice using something like NOTEPAD. That's it!!

Save the file then click on a HTR character to bring it up in the Haptek window.

Now locate your text file and do the DRAG & DROP onto the Haptek window. The character will read your text file and the lipsync works fine! If we could figure out how to change the voice it would be even better.

Oh yes, be sure to start your text file with a carriage return (Blank line or hit ENTER), then type what you want or else it won't work!

In the world of AI it's the thought that counts!

- Art -

Maviarab

Art,

Now that is interesting.

Again...the possibilities are endless here with a bit more fiddling around etc.
AIDreams...The new home of the Digital Girl

www.aidreams.co.uk

Duskrider

#2
You can pick your voice in Hal.
You do your drag and drop to Haptek console if you using HapLogCtrl.
Your Hal-gal will use whatever voice you have picked for her.  
I believe there's no limit to how much she can read.  
Type a story for the kids.  

And yes, Hal does lipsync.

Enjoy.
[;)]

Maviarab

#3
Jim,

Never mind typing stories lol...with the thousand of ebooks available on the net just convert those (as most in pdf) to txt and drop those in ?

if your right (not had time to test fully) that there is no file size limit then we can just get the haptek character to read an entire book...not sure on how spaces between paragraphs would work out though ???

Will have to test that in the morning. Will report back [:D]
AIDreams...The new home of the Digital Girl

www.aidreams.co.uk

freddy888

I've got the beginnings of an idea on this.

I've been picking the htr's apart and have been looking at the switches that produce the lipsync.  These can be called just like any other switch so in theory you should be able to write a piece of VBasic script that goes through a line picking out the speech patterns then calling the right mouth position.  That's what the commercial software would be doing.  There aren't that many as well.

I can't see why that wouldn't work inside Hal as an addon to the brain file.