The strength of each word is derived strictly from its position in the phrase (marked for the speaker with a hyphen), its position in the utterance, and it's position relative to the strongly accented word, or the word that the speaker is seeking confirmation. Specifically,
Then, x is the position of the word in its phrase, changing smoothly from zero at the beginning to 1 at the end of the phrase, and gives the slope that controls how the strength of words varies inside a phrase. Normally, < 0, to give phrasing with more emphasis (higher pitch) at the beginning; = 0 implies no phrasing.
Similarly, X is the position of the word in the utterance, changing smoothly from zero at the beginning to 1 at the end of the utterance, and gives the slope that controls how the strength of words changes across the utterance. Declination is equivalent to < 0, no declination to = 0, and a rising intonation can be obtained with > 0.
Finally, controls how strongly the phrasing is suppressed after the accent: = 0 means no phrasing or declination, and = 1 implies equally strong phrasing on both sides of the accent. Each of these parameters affects from tens to hundreds of words.
Boundary tones have a strength that is calculated from the strength of the first or last word (as appropriate):
sbt = . sw, | (9) |