Gösta Bruce

Textual aspects of prosody in Swedish.

The present paper explores the textual function of intonation in indicating coherence and boundaries between successive utterances of speech. An informal listening test showed how such prosodic coupling can be used to mark coherence between sentences of a text unit. The coherence is expressed as a downdrift of F0 peaks and valleys characterizing the whole text unit. The boundary between two text units is signalled by terminating one and beginning another downdrift ramp. The characteristic features of the downdrift pattern seem to be (1) an adjustment of F0 values to the length of the text unit, and (2) an adaptation of the F0 of the beginning of a succeeding sentence to that of the later part of a preceding one in the same unit. Some preliminary implications of these results for speech planning are discussed.

On the analysis of prosody in spontaneous speech with exemplification from Swedish and French

Abstract This paper reports on a research project concerning prosody in spontaneous spech. Two questions inaugurate the project. The first one concerns prosodic differences between spontaneous speech and read speech. Evidence from Swedish shows that these differences are not fundamental. The second question concerns the relationship between prosody and discourse categories. A methodology has been developed in order to study this relationship. Four different kinds of analyses are applied: (1) analysis of the discourse structure of the speech corpus without specific reference to prosodic information, (2) auditory analysis in the form of a prosody-oriented transcription, (3) acoustic-phonetic analysis and (4) analysis-by-synthesis. Part of this analysis is illustrated with exemplification from a persuasive monologue in French political rhetoric. Focal accent and contrast in pitch range seem to account for typical prosodic means used during political debate.

Prosodic modelling in Swedish speech synthesis

Abstract Our present work concerns Swedish prosody in a speech synthesis framework. Two main problem areas are examined: prominence and phrasing. In a model for Swedish prosody, prominence levels (stress, accent, focus) are represented as layered and multidimensional for different domains (syllable, foot, word). Phrasing involves both coherence in the form of specific combinations of existing accentual gestures and separate boundary gestures. The main features of the intonation model are given in outline. Experiments on prominence include modelling of durations in a combined speech data base and rule synthesis framework, where the stressed-unstressed alternation appears to be the most important duration factor. Other experimentation concerns typical differences in the timing characteristics of the tonal gesture for focal accent between compound words and simplex accent II words. Experiments on phrasing include both production data from a varied speech material as well as synthesis and perception. Our experiments demonstrate that both coherence and boundary cues are effective as phrasing signals and that a combination of F0 and duration is typically used to signal phrasing. Our future plans include working with prosodic modelling of Swedish in a dialogue context and in a concept-to-speech framework.

Cue words and the topic structure of spoken discourse: The case of Swedish men 'but'

The Swedish cue word men ‘but’ can mark the boundary between both different topic units as well as topic-internal units in spontaneous speech. The goal of this study is to see if these two functions of men can be distinguished on the basis of their local prosodic correlates and co-occurring lexical items. Men-tokens in spontaneous narrations were labelled as to their function, first using text-only data. The ‘strong’ tokens (categorized identically by all labellers) were subsequently seen to be clearly differentiated into two classes on the basis of related prosodic parameters and co-occurring lexical items. This distinction was, however, not found for the corresponding ‘weak’ tokens which were subsequently relabelled using both text and speech nor for the data-base as a whole. A test using a neural network trained using strong tokens was seen to be able to correctly categorize 90% of the strong men-tokens as to their associated boundary-type (topic-shift vs. topic-internal). The results show that cue words along with their prosodic correlates and co-occurring lexical items constitute a constellation of important information for understanding how segmentation of spoken discourse is produced and understood.

On the Analysis of Prosody in Interaction

The research reported here is conducted within the ongoing research project “Prosodic Segmentation and Structuring of Dialogue”. The object of study in the project is the prosody of dialogue in a language technology framework. The specific goal of our research is to increase our understanding of how the prosodic aspects of speech are exploited interactively in dialogue—the genuine environment for prosody—and on the basis of this increased knowledge to be able to create a more powerful prosody model. In this paper we give an overview of project design and methods and present some tentative findings.

Modelling of Swedish text and discourse intonation in a speech synthesis framework

In this chapter, we present a model for the analysis and synthesis of intonation in spontaneous conversations in Swedish. The model is an enhanced version of the model developed in (1977) and implemented in our text-to-speech synthesis. In our recent work we have developed the model from the perspective of discourse and multi-sentence texts. This takes us out of the restricted one-sentence/utterance analysis and synthesis into the living world of prosody in communication.

Developing the modelling of Swedish prosody in spontaneous dialogue

The main goal of our current research is the development of the Swedish prosody model. In our analysis of discourse and dialogue intonation, we are exploiting model-based resynthesis. By comparing synthesized default and fine-tuned pitch contours for the dialogues under study, we are able to isolate relevant intonation patterns. This analysis of intonation is related to an independent modelling of topic structure consisting of lexical-semantic analysis and text segmentation. Some results from our model-based acoustic analysis are presented, and its implementation in text-to-speech-synthesis is discussed.

The phonetic profile of Swedish

Abstract This chapter is “typological” in two senses: Whereas the first section considers some of the sounds and sound patterns of Swedish from a universal-typological point of view, the second section discusses the considerable phonetic variability observed across the various dialects of the language. It is argued that, with some exceptions, Swedish is typologically fairly mainstream. Exceptions concern particularly the inventory of non-back rounded vowels, voiceless fricatives and partly prosody. The Swedish dialects are found to contain several distinct consonant and vowel types that are not encountered in Standard varieties. It is also found that the intonational structure of the Swedish dialects is fairly complex and diversified. The third section concludes the chapter with some informal observations of possible sound changes in progress.

Revisiting Southern and Central Swedish intonation from a comparative and functional perspective

Revisiting South and Central Swedish intonation from a comparative and functional perspective

Aspects of Danish prosody

Abstract The volume Aspects of Danish Prosody presents a collection of papers originating within a research programme in Denmark in the late 80s and early 90s called Spoken Danish in its Variations. This research was initiated and sponsored by the Danish State Research Council for the Humanities (SHF) during the period 1986 – 92. One of the focal areas of this umbrella project represented in the present volume is the prosody of natural and spontaneous speech. The ultimate goal of this research effort has been to contribute to the knowledge of contemporary spoken language by exploiting (in particular) natural speech data as opposed to laboratory speech, which until quite recently has been the dominating type of speech data within the area. The main speech material used in the prosody project comes from another part of the research programme, dealing with sociolinguistic research, specifically based on interviews with inhabitants of a certain urban district of Copenhagen (cf. Frans Gregersen and Inge Lise P...