Compound noun phrases

The simplest prominence prediction algorithm simply uses the concatenation of the lexical prominence patterns of the words. So for  

RESULT WAS A CLEARLY A GOOD ONE FOR SHAW we might generate a prominence pattern of  

The main ways in which this rule fails (again ignoring affective and augmentative effects) are in compound noun phrases and certain function words uses. The prominence patterns in compound noun phrases in English are complex and can seem quite baffling. When we consider the prominence patterns in street names such as 

This principle can then be generalised to cope with all two word cases, with the idea being that only conventionalised cases are treated as compound words which receive their prominence on the first word. In addition to this, there are a number of semantically driven rules, which range 

For more compound noun phrases involving three or more words, a C YK parser (see Section 5.5) is used to assign the internal structure to the noun phrase, after which prominence values can be assigned. To resolve ambiguity, a set of heuristics is used to find the best parse. These are  

Prefer parses determined by semantic rather than syntactic means. (This works partly because semantic parses are much more specific.) 

---

Prominence patterns are particularly noticeable in sequences of adjectives and nouns which are collectively called compound noun phrases. For example... 

Prominence prediction by deterministic means is actually one of the most successful uses of non-statistical methods in speech synthesis. This can be attributed to a number of factors, for example the fact that the rules often don t interact or the fact that many of the rules are base on semantic features (such that even if we did use a data driven technique we would still have to come up with the semantic taxonomy by hand). Sproat notes [410] that statistical approaches have had only limited success as the issue (especially in compound noun phrases) is really one of breadth and not modelling regardless of how the prominence algorithm actually works, what it requires is a broad and exhaustive list of examples of compound nouns. Few complex generalisations are present (what machine learning algorithms are good at) and once presented with an example, the rules are not difficult to write by hand. 

Prominence is used to give extra strength to certain words. It is partly determined by syntax, where in particular it governs stress patterns in compound noun phrases and function word sequences. 

---