Using information about syntactic dependency

[Luca B]

Is it possible for a speech synthesis software to use linguistic information about syntactic dependency? If so, do you know of any resources I could look into regarding this topic?

I’m mainly interested in this for phrase break prediction – my intuition here is that when there is a dependency between, say, word A, B and C, then those words form a natural intonational phrase.

[Luca B]

If anyone else is interested in this, I just found a paper on this topic:

Mishra, T., Yeon-Jun Kim, & Bangalore. (2015). Intonational phrase break prediction for text-to-speech synthesis using dependency relations. 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015, 4919-4923.

[Author]

Posts