This table represents the values to set in the tag of the table it’s used to set Pepper’s voice:
|Voice height||\ Vct = 135 \||50 to 200|
|Voice speed||\ Rspd = 110 \||50 to 400|
|Weight||\ Pau = 1000 \||1000/1 second|
|Voice volume||\ Vol = 50 \||0 to 100|
The default settings makes quite hard to to listen to long sentences and the voice height is a bit too low, so when speaking quite a bit, the speed is around 90 or 100 and the height of the voice depends on the words. Normally I am using about 170 lines when I am happy with about 145 or when my tension is rising.
You may try setting the intonation only by the height of the voice, but I think that it will not work as you are expecting because words will be chopped when you set the voice height setting to shred words.
Here at this link http://doc.aldebaran.com/2-5/naoqi/audio/altexttospeech-tuto.html you can find more detailed information that covers also other customisation capabilities of ALTextToSpeech module.