Linguistic theories and prosody

Given these difficulties with labelling schemes and models, it is worth asking whether we need a model of prosodic form at all perhaps it would be possible to simply define a mapping from the text or semantics to the acoustics. This viewpoint certainly has some merit, but we should be wary of dismissing the idea of prosodic form entirely. A brief digression into the theoretical history of prosody may help explain why. 

It is probably fair to say that the field of linguistics has always had a troubled relationship with prosody. In fact many accounts of language completely ignore prosody (e.g. prosody is hardly ever mentioned in any work by Chomsky) whether this is because it is deemed to be quite separate from verbal language or simply too tricky to handle is hard to say. If we however want a full account of language and communication we have to include prosody. 

The consequence of this, taken in conjunction with the issne of a very low level of transcribed agreement, may well mean that many prediction algorithms are as accnrate as theoretically possible no amount of extra refinement in the techniqnes or extraction of alternative features will make much difference. The reason why the algorithms cannot improve is due to noise in the data, from either or both of these sources. 

---