Synthesis with superimpositional models

The Fujisaki model is most commonly used with Japanese, but has been used or adapted to many other languages. In Japanese, we find that the range of pitch accent phenomena is narrower than in languages such as English, which means that the model s single type of accent is particularly suited. In addition, the nature of intonation in Japanese means that accents are marked in the lexicon, which greatly simplifies the problem of prominence prediction. Hence a simple approach to this, which uses accent information from the lexicon alone, is often sufficient. A common approach therefore is to determine phrase breaks and prominent syllables from the text, and then phrase by phrase and syllable by syllable generate the input command parameters for the Fujisaki 

One significant advantage that Hirose makes note of is that even when the prediction algorithm generates incorrect parameters for the Fujisaki model, the result is still a natural contour because the model itself is tightly constrained. In other words, the contour produced is some contour, just not the right one. By comparison, when prediction is poor with target models, the result may be a contour that bears no similarity to the fype that a human could produce. 

---