Linguistic and Acoustic features

Both the target and the join function operate on a feature structure of the units and specification, and conveniently we can use the feature-structme formalism for both. The Hunt and Black framework doesn t limit the type or nature of the features or their values in any way, so in principle we are able to use any features we want. 

---

The specification and the units are completely described by a feature structure, which can be any mixture of linguistic and acoustic features. 

Structural linguistics deals with how langui es are built and ordered into rules for communication. This subdiscipline, a basis for scientific research, clinical studies, sociological explorations, or scholarly pursuits, focuses on how speech is physically produced. Structural linguistics involves acoustic features, the comparison of different languages, grammatical concepts, sentence construction, vocabulary function, and other analytical aspects of language. 

One of the fundamental problems in unit selection is that the specification items lack the acoustic description(s) that would make matching them with units a fairly easy process. We can approach this problem in two ways. Firstly, we can just ignore the fact that the units have acoustic features and just match on the linguistic features alone. Alternatively, we can try and perform a partial synthesis, where we attempt to generate some or all of the acoustic features and then match these with the acoustic features derived by signal processing fi-om the waveforms. 

The join feature issue is considerably more straightforward than with target features. Firstly, we have access to both linguistic features and the true acoustic features secondly as all units have had their features measured and derived in the same way, we don t have to worry about issues of comparing generated acoustic values with ones determined by signal processing. In addition to... 

As well shall see in Sections 15.2.4 and 16.4.1, decision tree clustering is a key component of a number of s mthesis systems. In these, the decision is not seen so much as a clustering algorithm, but rather as a mapping or function from the discrete feature space to the acoustic space. As this is fully defined for every possible feature combination, it provides a general mechanism for generating acoustic representations from linguistic ones. 

A linguistic join cost can be defined which looks at the differences between categorical/linguistic features across a join. Acoustic and linguistic join costs cab be combined in a number of... 

---