Speech tempo and phonetic reduction in German

• Applicant: Professor Dr. Beat Siebenhaar
• Subject Area: Individual Linguistics, Historical Linguistics
• Term: from 2017 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 383554615

Speech tempo and phonetic reduction are regionally different in German (Hahn/Siebenhaar 2016). This issue is explored on the basis of 668 recordings of the reading text 'The North Wind and the Sun' from 167 locations in German-speaking Europe. The recordings were collected between 2006 and 2009 in the context of the 'Deutsch heute' corpus at the Institut für Deutsche Sprache (Mannheim). Upon the applicant's request, the speakers read the material in 'normal' and 'fast' reading speeds.The aim of this research project is to clarify whether speech tempo in the German standard varies regionally, whether there are areas with the same intrinsic speech rate, whether there exist regionally specific phonetic reduction patterns, and ultimately we intend to show if and how both factors, speech tempo and phonetic reduction, potentially covary in space. Therefore, popular views on where people speak fast or slowly, clearly or blurredly can be set on an empirical basis. This monodimensional description is followed by a structured multilevel explanation. The two speech tempo 'normal' and 'fast' are seen as two stages in a continuum. They both are individually evaluated and compared, since it can be assumed that, in the case of rapid speech, regional preferences of pronunciation are more prominent. This approach makes it possible to distinguish lexicalized 'reducts' from of tempo-related articulatory recent 'reductions'. The analysis of the differences between the two stages thus provides a dynamic view, which reveals the regional 'crumbles', ie the strategies of phonetic savings that are necessary to speak 'faster', which remain unobservable if one does not compare the two speech tempi.Articulatory speech rate and rhythm measurements make it possible to roughly organize the data. However, these general measures only give a limited indication of how differences are produced and how different phonetic factors correlate. To determine this, the duration of individual sounds is examined for their variation. In addition, aspects of sound quality are taken into account in order, for example, to check the reduction of the vowel space or the reduction of the voicing of obstructions with regard to regional-specific preferences. Moreover, phonetic simplifications (eg elision and assimilation) are recorded. Additionally, statistical factor analyses allow for a reduction of the data to factors that can be attributed to those variables which account for the main speech rate or reduction variance. These resulting factors are geostatistically analyzed, mapped and interpreted, whereby the determined spatial structures provide new insights into traditional language geographic classifications.

• DFG Programme: Research Grants
• Cooperation Partners: Privatdozent Dr. Christoph Draxler; Dr. Stefan Kleiner; Dr. Gudrun Mayer