Acoustic phonetics

We know fi om acoustic phonetics that typical formant values for an /ih/ vowel are 300Hz, 2200Hz, and 3000Hz. Formant bandwidths are harder to measure accurately, but less us assume a value of 250Hz for all three formants. Assuming a sampling frequency of Fs = 16000/7z, the following table shows how to calculate the poles from the formant frequencies and bandwidths. 

I will finish, though, by saying that a solid knowledge of the basics of linguistics has been invaluable to me in my own TTS research, and that, in my view, a solid knowledge of acoustic phonetics is one of the key skills required to build a good unit-selection... 

Prosodic phrasing is to some extent governed by purely phonological, phonetic or acoustic factors which can override the syntax. In the classic example, Chomsky [89] commented on the fact that in... 

It is difficult to discuss intonation further without reference to the phonetics and acoustics of this part of language, and for that reason we forgo a more thorough account of intonation until Chapter 9. 

In traditional phonetics, phones were seen as a set of distinct, discrete, separable units. This notion is no longer defensible as we now know that in reality the phonetic space is a multidimensional continuum (describable either with articulatory or acoustic dimensions), and there are no bottom up divisions within it. The concept of using phones should now be seen as an idealised division of phonetic space and this can be justifiable as a handy notational device it is cumbersome to describe every sound in terms of continuous dimensions (of say tongue height). It is important to realise however, that phones really are a discrete abstraction of a continuous space no procedure or process can be given for objectively defining these units. 

The most popular work in introductory phonetics is undoubtedly Peter Ladefoged s classic a course in phonetics [271], and is highly recommended. Another excellent book, which extends the subject and gives more detail on acoustic issues is Clark and Yallop [97]. A more advanced and comprehensive book is Laver s Principles of Phonetics [275]. 

The models differ significantly in what they take as the primary form of intonation. In the AM model this is quite abstract while in the Tilt model this is quite literal or acoustie . These differences in primary form should not be taken to mean that the proponents of these models do not believe that there should be more abstract or more conerete representations, just that the best representation happens to lie where they describe it. In the many synthesis schemes based on the AM model there are other, more phonetic or acoustic levels, and in the Tilt model there is always the intention that it should serve as the phonetie description of some more abstract higher level representation. 

Elovitz, H. S., Johnson, R., McHugh, A., and Shore, J. Letter-to-sound rules for automatic translation of English text to phonetics. IEEE Transactions acoustics speech and signal processing 24 (1976), 446-459. 

Paulo, S., and Oliveira, L. C. Dtw-based phonetic alignment using multiple acoustic features. In Proceedings of Eurospeech 2003 (2003). 

---