Hi guys(/girls), I am currently a Master student and my research interest is on speech processing particularly in Articulatory Synthesis area, but I kind of interested in image processing too.

Therefore, I want to ask any of you that has any experience with both of those areas, as I wanted to integrate them together, such as a talking 3D head (from my readings) or any other that you may have conducted research before or anything.

This is an 'opinion asking' thread, not a 'do my homework' thread.


Thank you for reading.

Sorry I cannot picture your problem there.. How are you going to integrate image processing?

Do you mean something like lip reading?

Yes, something in that area :-), basically what I have in mind right now is to detect the image of the lip/mouth, then transform it to speech.

Orite.. I got it.
I guess you need some image processing library like OpenCV
and some algorithm like
Lucas Kanade and CAMShift / Meanshift for tracking
some neural net for classification

even human cannot read other people's lips...

Lip reading may not be successful with mono syllable language.

Thanks Mr. SanosukeSagara for that informative reply

I think this is a potential project I see quite a few successful implementation in publication and paper.

another important part might be huge database of classification.