I am going to build a system which will produce non-verbal behavior in a data-driven way, based on the speech signal and its transcription.
I need to decide which speech feature to use, so that I can train it on the human recordings and then use it on the humanoid robot NAO.
Main problem is that the robot's speech will not have the variability of the natural speech,as it is produced by the text-to-speech system. So I need to be carefull in not learning smth, that can work only for humans and will not work on my robot.
As Mohammed said it already, the Fundamental Frequency (F_0) is the key feature to design such a system to produce a "natural" speech by Speech Synthesis. You can also add other features like the two first Formants.
My question was more about speech processing rather then producing. Main thing I want to produce is the motion (not the speech) and I want to do it based on the speech. I will train my system on the human recordings.
So it is important that what I will extract from the human speech will be the same for the robot speech as well.
Mohammed and Kofi, you are saying that F0 is invariant to the speaker/robot production?
- Badges
- Science topic