Linguistically motivated parameter estimation methods for a superpositional intonation model EURASIP Journal on Audio, Speech, and Music Processing, Volume 2014, Issue 1.

TORRES HM..; GURLEKIAN JA..; MIXDORFF H.. ; PFISTZINGER H..

EURASIP JOURNAL ON AUDIO, SPEECH AND MUSIC PROCESSING

HINDAWI PUBLISHING CORPORATION

Lugar: New York; Año: 2014 vol. 1

This paper proposes two novel approaches for parameter estimation of a super-positional intonation model. These approaches present linguistic and para-linguistic assumptions for initializing a pre-existing standard method. In addition all restrictions on the configuration of commands were eliminated. The proposed linguistic hypotheses can be based on either pitch accents or lexical stress, which give rise to two different estimation methods. These two hypotheses were validated by comparison of the estimation performance relative to two standard methods, one manual and one automatic. The results of the experiments for German, English and Spanish corpora show that the proposed methods outperform the standard ones.