Klaus J. Kohler

Segmental Reduction in Connected Speech in German: Phonological Facts and Phonetic Explanations

This paper deals with /r/ vocalization, weak forms, elisions and assimilations in connected speech in German and provides phonological rules within the generative framework for the derivation of any observable reduction forms, given certain phonetic environments. Some syntactic constraints are also discussed. This phonological description is then supplemented by an enquiry into the production and perception constraints of real speakers and real listeners to find answers to the questions as to why certain articulatory modifications occur and others do not, and why they take particular directions. This phonetic explanation is sought in a minimization of energy expenditure in relation to the perceptual demands of the communicative situation.

Phonetic Explanation in Phonology: The Feature Fortis/Lenis

This paper examines the wide spectrum of phonetic properties associated with the phonological distinctions between consonant classes such as /p, t, k/ and /b, d, g/ in a great variety of languages (including languages with multivalued contrasts) and evaluates their relationship to the features [+/- voiced], [+/- aspirated], [+/- fortis]. The discussion separates word-initial, word-medial and word-final positions according to their different production and signalling constraints. A power feature, realized in articulatory timing and/or phonatory power/tension, is set up as the essential differentiator, thus providing a phonetic basis for the fortis/lenis dichotomy. The incorporation of the time dimension into phonology is regarded as a necessary prerequisite to the solution of phonological problems in general and to an adequate treatment of the [+/- voiced] feature in particular, voice onset time being only one temporal aspect.

Glottal Stops and Glottalization in German

This paper presents the results of an auditory and instrumental investigation into glottal stops and glottalization as boundary markers in German colloquial read speech of a North German non-dialect v

Rhythm in speech and language: a new research paradigm.

Like any other aspect of spoken language, rhythm needs to be, and has been, studied from four different perspectives for a comprehensive and insightful account of its nature and functioning in speech communication: symbolic representation, production, perception, communicative function. The paper first gives an overview of the milestones in the analysis of rhythm under the headings of these four approaches over the past 70 years. This survey of the development of scientific ideas in rhythm research prepares the ground for the integration of the four strands in an interrelated framework of linguistic and speech signal analysis. On the basis of a definition of rhythm derived from the theoretical and methodological discussion, a new paradigm is outlined for future research, centred on the listener and on communicative function.

Timing and Communicative Functions of Pitch Contours

A new research paradigm is applied to F0 synchronization with articulation, inpeak and valley contours, under four principles: (a) Timing of F0 contours enters thedefinitions of the pitch categories. (b) These phonetic exponents are linked to com-municativefunctions. (c) The listener plays a pivotal role. (d) Contextualization oftest stimuli is essential for pitch data collection. Data are presented and interpretedfrom an experimental investigation of the substance-function relationship in theperception of peak and valley shift series by German listeners, using the semanticdifferential technique. The findings of the substance-function relationship areexplained with reference to the frequency codeand to auditory enhancement,i.e.syntagmatic contrast of high-low or low-high pitch trajectories in synchronizationwith articulatory landmarks of accented syllables.

The Perception of Prominence Patterns

The term ‘stress’ is used to refer to the perceptual salience at certain places in strings of syllables, but it has several different referents: (a) relative syllable salience in an utterance; this is syllable-, not word-oriented; (b) stress in a word; this is part of the lexical phonology; (c) stressing of words in utterances for various aspects of propositional and expressive meaning, often called ‘accent(uation)’. Referents b and c are word- and meaning-oriented. In this article, the terms are more stringently defined. ‘Stress’ is only used to refer to a lexical stress position (referent b), i.e. a syllable in a word that becomes the docking place for various types of ‘accent’ to weight words in utterances (referent c). ‘Stress’ has no physical attributes by itself. ‘Prominence’ refers to the patterns of salience in syllable strings (referent a). The article reports results of an experiment in prominence perception of the logatome baba, in which the physical parameters F0, syllabic duration, and overall acoustic energy were systematically varied across the bisyllable. Sixteen German subjects had to indicate, by pressing buttons of a computerized reaction time device, whether the first or the second syllable was more prominent. F0 was a more powerful cue than the other two. Equal syllable duration on a monotone resulted in more first-syllable judgements, which could be counteracted by a slightly falling F0 contour on the second syllable to reach equal response frequencies for the two syllables. This ties in with Lehiste’s earlier findings that F0 movement increases the perception of duration. Extrapolating from the results, a research programme for prominence perception is developed that will eventually shed new light on the investigation into the nature and manifestation of speech rhythm.

Modelling Prosody in Spontaneous Speech

Following on from general considerations of requirements for prosodie modelling of spontaneous speech, this paper outlines a prosody model for German, its incorporation in a TTS system as a prosody research tool, the model-based development of a prosodie labelling system for the application to spontaneous speech, and the use of the resulting prosodie label files as input to the TTS system for transcription verification and model elaboration.

Labelled data bank of spoken standard German: the Kiel corpus of read/spontaneous speech

The structural frame for a computer databank of spoken German and its integration into basic and applied phonetic research has been developed at Kiel University in Germany (IPDS) and is continually being filled with segmentally and prosodically labelled data from read and spontaneous speech recordings. In future more diverse types of spontaneous interactions will be included in the Kiel Corpus. The range of phonetic analyses will be extended, also to include prosodic variables and comparisons of speaking styles. Databank analyses will provide rules for connected speech processes, which in turn will assist in the development of automatic segmentation and labelling procedures. Finally, this phonetic databank concept can be extended to other languages to initiate large scale multilingual phonetic comparisons.

Investigating Unscripted Speech: Implications for Phonetics and Phonology

This paper looks at patterns of reduction and elaboration in speech production, taking the phenomenon of plosive-related glottalization in German spontaneous speech, on the basis of the ‘Kiel Corpus’, as its point of departure, and proposes general principles of human speech to explain them. This is followed by an enquiry into the nature of a production-perception link, based on complementary data from perceptual experiments. A hypothesis is put forward as to how listeners cope with the enormous phonetic variability of spoken language and how this ability may be acquired. Finally, the need for a new paradigm of phonetic analysis and phonological systematization is stressed, as a prerequisite to dealing adequately and in an insightful way with the production and perception of spontaneous speech.

F0 in the perception of lenis and fortis plosives

This paper reviews articulatory data that provide evidence for a backward influence of stop manner production on the F0 of the preceding vowel in German, and it presents results from four experiments that show the relevance of pre‐closure F0 for stop manner perception in the German word pair ‘‘leiden’’/‘‘leiten.’’ The results are compared with several perceptual investigations of monotone versus changing F0 on English monosyllables influencing the voiced/voiceless judgment for post‐vocalic consonants. The theory that we are here dealing with a general auditory phenomenon instead of a production–perception relationship is refuted on the basis of the production and perception data available for German and for English.