A system of stylized intonation contours in German

Abstract

Modeling intonation, i.e., specifying adequate fundamental frequency (F0) contours, remains a challenging task for speech synthesis systems. This paper discusses the development of a system for phonetically specifying intonation contours for German. It deals with the problem of translating an abstract phonological representation of intonation namely the tone-sequence model into a concrete phonetic model. Design options and evaluation methods are discussed. 1. PHONOLOGICAL SPECIFICATION The study of phonological aspects of intonation has strongly been influenced by the tone-sequence model of Pierrehumbert [12]. As a consequence, the ToBI (Tone and Break Indices, [3]) notation system has become a generally accepted means to describe intonation structure phonologically. In this autosegmental model intonation contours are modelled by specifying a sequence of high (H) and low (L) target tones. In [13] a ToBI offspring for the description of German intonation is presented. Neglecting details like phrase accents and downstepping for now, this system comprises the following inventory of tones: H* Normal peak accent L+H* Steep rise within the accented syllable L*+H Peak is moved behind the accented syllable L* Valley accent H+L* Falling accent L% Low boundary tone H% High boundary tone