this question is uncategorized.. i guess.
I wonder can i convert every recorded voice to one standard voice to be able operating on it?
Let me explain on image. I have colored image but i know that it’s very hard to recognize letter. So lets convert it. I can now operate on image which is in grayscale. I have more abilities to work on it then.
Few days i finished my own alghoritm recognizing letters in the picture, so i have some experience. :]
Can you give me some good ideas or alghoritms about converting voice to voice on which i can work easily?
edit1
lets say that there are two people talking to my pc’s mic. Look that the graph of my voice and my friend voice are different. I ask, is it any possibility to normalize both voices to one universal voice. If i program “Hi! – by me, it won’t work if my girlfirend say “Hi!” in my program (written in c# ).
Based on your update I think you’re looking for a combination of voice recognition software and speech synthesis .
I suggest looking at CMUSphinx: http://cmusphinx.sourceforge.net/
and festival: http://festvox.org/
respectively.
So use voice recognition to determine what people are saying and output as a standard “voice” using speech synthesis.