Since the 14th International Congress of Phonetic Sciences in San Francisco in August 1999, two decisive events have set signals for the future directions of phonetics in general and of the International Phonetic Association in particular. First of all, the Permanent Council for the Organization of International Congresses of Phonetic Sciences, the principal, quadrennial international forums for the presentation of phonetic research, and the Council of the International Phonetic Association, the oldest and most prominent scientific society of phonetics, separately voted in favour of a union, with the intention to affiliate the Permanent Council as a standing subcommittee to the IPA Council and to run future Congresses under the auspices of the IPA. This will broaden and intensify the activities of the IPA in all areas of Phonetic Science, even if the phonetic descriptions of languages will remain a traditional focus of attention. Secondly, a decision was taken by the IPA Executive to reach agreement for the Association's Journal to be published by Cambridge University Press. Upon the conclusion of the contract, we can now proudly present the first issue of volume 31 of the Journal of the International Phonetic Association under its new aegis. In conjunction, the two decisions taken by the IPA open up the prospect of a powerful international platform for the distribution of the state-of-the-art and new results in phonetic research.

This paper provides a statistical account of schwa elision and vowel nasalization, and of nasalization and deletion of plosives in a large corpus of German spontaneous dialogues in comparison with an equally large data base of read speech (sentences and texts) from large groups of North German speakers. The phonetic variability of these phrase-level processes is projected onto the articulatory dynamics in global opening and closing gestures, which are taken to be basic phonetic structures of speech communication. Trends for gesture reorganization are derived from statistics, and related to external control factors of word boundary, word class, speech style as well as internal phonetic conditions of gestural make-up and of reduction of articulatory complexity. These synchronic facts of one language are compared with parallel instances from other languages and linked to congruent diachronic data of sound change, thus laying the foundation for generalizable phrase-level patterns of human speech production.

Labellers of connected speech corpora encounter a number of cases which are difficult to handle with a linear sequence of segmental labels. Based on material from a corpus of German unscripted speech, this paper presents non-sequential aspects of laryngeal activity for /x/ and /h/, and of supralaryngeal articulation in vocoid exponents of postvocalic /r/. Comparative data for /r/ from a corpus of German read speech indicate that differences across speaking styles may be independent of durational variation. It is discussed how models of the phonology-phonetics interface and Prosodic Analysis can account for the findings.

Articulatory explanations are often proposed to account for many of the phonetic patterns found in speech beyond the citation form. Unscripted material from Suffolk English and North German is used to argue that articulatory explanation can too tightly constrain our expectations of the types of phonetic patterns which spontaneous speech contains. It is also shown how articulatory explanation can offer an adequate account of a dataset, stopping short of revealing a larger, more complex set of phonetic patterns.

Informal listening suggests that unscripted Swedish shows a tendency to produce alternating contoid and vocoid articulations which relate to more complex consonant and vowel structures at the phonological level. To test this hypothesis, two unscripted monologues and, for comparison, a careful text reading were analyzed. The speech material was segmented using criteria based on the so-called sonority hierarchy. The results largely corroborated the hypothesis in showing that contoid-vocoid units appeared considerably more frequently in unscripted speech than suggested by conventional phonotactic analysis, and that some reduction effects appeared more frequently in the unscripted than in the read speech. The possibility that this reflects an underlying articulatory organization of spontaneous speech in terms of typologically basic CV units is discussed.

Quasi-spontaneous dialogues from six languages which, according to recent discussion of rhythmic types, belong to three rhythmic groups – Russian and Bulgarian as ‘stress-timed’, Italian and Greek as ‘syllable-timed’ and Polish and Czech as an intermediate ‘mixed’ type – were examined for the following segmental reduction phenomena: reduction of consonant clusters, weakening of consonant articulation, residual properties from elided consonants in the original context segments, phonetic schwa-isation and syllable elision. The hypothesis tested was that there are comparable reduction phenomena in all languages since all languages allow for variation in the time and effort invested in any given part of an utterance as a means to support the relative weight of elements within the information structure. This hypothesis was borne out in principle, though there were a small number of exceptions across the six languages to the occurrence of reduction types examined.

This paper aims to examine spatio-temporal variability during the production of the lingual consonants /t, d, s, z, k, g l, n, XXX/ in a sample of Greek spontaneous speech. It provides a description of the range of segmental variability and explores the relationship between articulatory and temporal variability. It also discusses instances of common connected speech processes including assimilations and deletions. Data from two Greek speakers producing a monologue was recorded with the technique of electropalatography (Reading EPG system). The results indicated that segmental variability ranges over a continuum from over-articulated to under-articulated forms, i.e. fully articulated tokens to very open articulations. In addition, duration dependent variability was found to be a function of the production characteristics of the consonant. Evidence of both gradient and categorical changes in spontaneous speech were suggested by the data and are discussed with reference to current theoretical frameworks.

The goal of this paper is to compare the degree of anticipatory and carryover phenomena in the behaviour of the velum during the production of French spontaneous speech, and to compare the data with the same speech material, but read. Airflows through the nose and the mouth were taken as an indirect indication of the velum behaviour. French is a language of special interest because there is a phonological contrast between oral and nasal vowels. The results show a tendency for nasal airflow to start before the nasal and a strong propensity to spread after the nasal. No differences could be established between speaking styles (spontaneous vs. read speech) regarding nasal airflow anticipation and carryover.

The present study was designed to test the effects of preceding and following context on the restoration of deleted and assimilated consonant sequences in conversational French speech. Consonant restoration was found to be enhanced by lexical information and by the different sources of information contained in the breath group. The fundamental contribution of breath-group information suggests that the breath group, which is a production unit in conversational speech, might be a perception unit. Interestingly, the results also appear to be in line with Lindblom's claim (1990, 1996) that speakers adapt their utterances to the listeners' needs and maintain sufficient contrast for promoting correct lexical access.

This paper presents data from a small corpus of unscripted speech gathered from one male and one female adult native speaker of Hindi. An acoustic analysis of the data demonstrated changes such as lenition and assimilation. The cases of assimilation included stop plus stop sequences yielding geminates. The results of a perceptual test showed that such ‘pseudogeminates’ are generally perceived as true geminates. Parallels between these phenomena and historical sound changes in Indo-Aryan are also discussed.

In spontaneous speech, occurrences of the same word form vary phonetically. Words that do not receive sentence accent tend to be less clearly articulated and shorter in duration in comparison to prominent words. If such an unaccented word is isolated from continuous speech, it may not even be identifiable by listeners, but as part of the speech stream, it is usually easily identified and helps the listener to perceive and organize the spoken message as a whole. Thus, the variability of unaccented words is not random but must follow certain patterns determined by context. In this study, we take a closer look at the Finnish unaccented word-like construction niinku in spontaneous conversational speech samples collected from one female Finnish speaker. We provide an analysis of the phonetic variability of the niinku structure found in the data, including such properties as segmental duration, segmental make-up, and phonetic quality. We then show what kind of properties are common to all occurrences of this unaccented construction and discuss the conditions that probably govern its acoustic-phonetic quality.

Finnish talk-in-interaction is shown to use creak and glottal stops distinctively. Creak has turn-yielding functions, and glottal stops have turn-holding functions. Rather than either intuition or the use of large corpora with no attention to the interactional function in which the talk is embedded, the methodology used is that of interactional linguistics (e.g. Couper-Kuhlen & Selting 1996 for a prosodic approach), which places emphasis on demonstrating participants' local orientation to linguistic categories within interactional sequences.

Unlike read or laboratory speech, spontaneous speech contains high rates of disfluencies (e.g. repetitions, repairs, filled pauses, false starts). This paper aims to promote ‘disfluency awareness’ especially in the field of phonetics –which has much to offer in the way of increasing our understanding of these phenomena. Two broad claims are made, based on analyses of disfluencies in different corpora of spontaneous American English speech. First, an Ecology Claim suggests that disfluencies are related to aspects of the speaking environments in which they arise. The claim is supported by evidence from task effects, location analyses, speaker effects and sociolinguistic effects. Second, an Acoustics Claim argues that disfluency has consequences for phonetic and prosodic aspects of speech that are not represented in the speech patterns of laboratory speech. Such effects include modifications in segment durations, intonation, voice quality, vowel quality and coarticulation patterns. The ecological and acoustic evidence provide insights about human language production in real-world contexts. Such evidence can also guide methods for the processing of spontaneous speech in automatic speech recognition applications.