User design artificial-voice technology – NTT media intelligence research institute

User design artificial-voice technology – NTT media intelligence research institute

It is NTT media intelligence research institute.
I suggest the service of the electronic book using the next-generation artificial-voice technology.
As a next-generation new technique, there are two features.
First feature is,
The synthetic sound can compose various people’s voice.
I can compose it only by one or two words talking in our technique.
I can compose anyone’s voice.
Thus far, the synthetic sound was able to compose it only in the voice of the authorized talent and voice actor.
However, the voice of me and voice of father, mother, grandmother and friend.
I can make the synthetic sound of wide contents.
Another characteristic,
The conventional voice synthesis was monotonous.
I was not able to reproduce enough intonations and power of expression.
However, this system extract an intonation from the voice of a narrator and the voice actor whom I recorded beforehand.
I put my voice and voice of grandfather on the top.
It can realize a high composition sound of the power of expression in the voice of various people.
When they put two characteristics together, contents like this are enabled.
I read Momotaro aloud. (A fairy tale of Japan)
Once upon a time there were grandfather and grandmother in a certain place.
Grandfather went to take the firewood to the mountain, and grandmother went to the river for washing every day.
One big peach flowed from the upper part of a river
DONBURAKOKKOSUKKOKKO
DONBURAKOKKOSUKKOKKO
*The onomatopoeia which is famous for DONBURAKOKKOSUKKOKKO – Momotaro.
How about it.
An intonation of “DONBURAKOKKOSUKKOKKO” is very natural.
The natural synthetic sound is not to have been able to readily do it so far.
I can realize the service of a new voice synthesis by putting the voice of my voice and your voice, the voice of the friend.
Thank you.