AI & NLP involving musical linguistics
 

Application of ‘Physics of Music’ to ancient Tamil grammar text ‘tholkAppiam’ and other ancient Tamil texts led to the discovery of ‘Musical Linguistics’ (‘Musical Phonetics in Tholkappiam’ in the journal of the international institute of Tamil studies; December 2013  http://www.ulakaththamizh.in/journal)

From the above discovery, the sound of letters had dual aspects; one serving the semantic goal in prose and speech, and the other serving the non-semantic musical goal in musically rendered song.

While linguistics is the scientific study of language, musical linguistics is the scientific study of the musical language employed in the musically rendered poems.

A poem can be rendered as a prosody or musical. In a musically rendered poem, musical structure-based encoding & decoding conveys the musical meaning. while simultaneous speech structure-based encoding & decoding conveys the language meaning.

In Musical Linguistics, the rules of joining the letters for composing the poems are non-semantic and hence language-independent, but musical structure dependent.

1.   Text to speech applications are based on the linguistics of the language used.

 

2.  Musically rendered poem of a language followed a set of rules unique to that language.

 

3.  The rules may govern the employment of letters, words, line and form of the poem.

 

4.  Language-independent, but the musical structure dependent universal grammar for musically rendered poems was discovered from the ancient Tamil grammar text tholkAppiam.

 

5.   The above discovery will be related to the emerging field of ‘Musical Linguistics’.

 

6.  Using a musically rendered poetic text as input, text to music application could be developed using AI & NLP, involving musical linguistics

Note: ‘Is ‘Chomsky’s thinking since at least 2002 is not really compatible with the practice of mainstream generative grammar’?’; https://www.academia.edu/s/79ef3bc4a9