Proceedings from FONETIK 2014 Stockholm, June 9-11, 2014
Total Page:16
File Type:pdf, Size:1020Kb
Proceedings from FONETIK 2014 Stockholm, June 9-11, 2014 PERILUS XXIV, June 2014 Edited by Mattias Heldner © The Authors and the Department of Linguistics, Stockholm University, 2014 ISSN 0282-6690 ISBN 978-91-637-5662-7 printed version ISBN 978-91-637-5663-4 web version Printed by US-AB, Stockholm 2014 Distributor: Department of Linguistics, Stockholm University Dedication This conference is jointly dedicated to Professor Björn Lindblom and Professor Anders Eriksson on the occasion of their 150th birthday June 19, 2014. The dedication is weighted with 53.3% to the former, and 46.7% to the latter of these. You phoneticians do the math. Previous Swedish Phonetic Conferences I 1986 Uppsala University II 1988 Lund University III 1989 KTH Stockholm IV 1990 Umeå University (Lövånger) V 1991 Stockholm University VI 1992 Chalmers and Göteborg University VII 1993 Uppsala University VIII 1994 Lund University (Höör) - 1995 (XIIIth ICPhS in Stockholm) IX 1996 KTH Stockholm (Nässlingen) X 1997 Umeå University XI 1998 Stockholm University XII 1999 Göteborg University XIII 2000 Skövde University College XIV 2001 Lund University XV 2002 KTH Stockholm XVI 2003 Umeå University (Lövånger) XVII 2004 Stockholm University XVIII 2005 Göteborg University XIX 2006 Lund University XX 2007 KTH Stockholm XXI 2008 Göteborg University XXII 2009 Stockholm University XXIII 2010 Lund University XXIV 2011 KTH Stockholm XXV 2012 University of Gothenburg XXVI 2013 Linköping University Preface This volume contains the contributions to FONETIK 2014, the XXVIIth Swedish Phonetics Conference, organized by the Department of Linguistics at Stockholm University, June 9-11 2014. The papers appear in the order in which they were given at the conference. Only a limited number of copies of this publication have been printed for distribution among the authors and those attending the conference. For ac- cess to electronic versions of the contributions, please look under http://www.ling.su.se/fonetik2014. We would like to thank all contributors to the Proceedings. We would also like to thank Språkstudion at Stockholm University for hosting the social event on the first evening of the conference. Finally, we would like to thank Fonetikstiftelsen for financial support. Stockholm in May 2014 On behalf of the Department of Linguistics Mattias Heldner, Francisco Lacerda, Marcin Włodarczak, Iris-Corinna Schwarz, Elisabet Eir Cortes, Hatice Zora, Lena Renner, Gláucia Laís Salomão, Ulla Bjursäter v Contents Session 1: Voice quality and emotion Voices after midnight – How a night out affects voice quality 1 Alexandra Berger, Rosanna Hedström Lindenhäll, Mattias Heldner, Sofia Karlsson, Sarah Nyberg Pergament, Ivan Vojnovic Emotional Finnish Speech: Evidence from Automatic Classification 5 Experiments Juhani Toivanen The intonation’s effects on speech intelligibility and attitudes 9 Sara Marklund, Jesper Zackariasson Session 2: Segmental features Instability in simple speech motor sequences - an overview of 15 measures and what they really quantify Fredrik Karlsson Tongue articulation dynamics of /iː, yː, ʉ̟ ː/ in Stockholm, Gothen- 17 burg and Malmöhus Swedish Susanne Schötz, Johan Frid, Lars Gustafsson, Anders Löfqvist An acoustic study of the Estonian Swedish lateral [ɬ] 23 Susanne Schötz, Francis Nolan, Eva Liina Asu Posters A data-driven approach to detection of interruptions in human– 29 human conversations Raveesh Meena, Saeed Dabbaghchian, Kalin Stefanov The WaveSurfer Automatic Speech Recognition Plugin 33 Giampiero Salvi, Niklas Vanhainen vi Towards a contingent anticipatory infant hearing test using eye- 35 tracking Iris-Corinna Schwarz, Atena Nazem, Sofia Olsson, Ellen Marklund, Inger Uhlén Session 3: Dialogue Duration and pitch in perception of turn transition by Swedish and 41 English listeners Margaret Zellers Backchannels and breathing 47 Kätlin Aare, Marcin Włodarczak, Mattias Heldner Pauses and resumptions in human and in computer speech 53 Jens Edlund, Fredrik Edelstam, Joakim Gustafson Session 4: Tone and accent Initiality accent deaccenting 59 Sara Myrberg Syllable structure and tonal representation: revisiting focal Accent II 65 in Swedish Antonis Botinis, Gilbert Ambrazaitis, Johan Frid Prosodic boundaries and discourse structure in Kammu 71 Anastasia M Karlsson, Jan-Olof Svantesson, David House Tonal production and syllabification in Greek 77 Antonis Botinis, Elina Nirgianaki Session 5: Animal sounds Sound initiation and source types in human imitations of sounds 83 Pétur Helgason Human perception of intonation in domestic cat meows 89 Susanne Schötz, Joost van de Weijer A pilot study of human perception of emotions from domestic cat 95 vocalizations Susanne Schötz vii Session 6: L2 Aspects of second language speech prosody: data from research in 101 progress Juhani Toivanen Perception and production of Swedish word accents by Somali L1 105 speakers Anna Hed The confusing final stops in L2 acquisition 111 Elisabeth Zetterholm Observed pronunciation features in Swedish L2 produced by two L1- 117 speakers of Vietnamese Mechtild Tronnier, Elisabeth Zetterholm Session 7: Child speech Consonant inventory of Swedish speaking 24-month-olds: A cross- 123 sectional study Emilie Gardin, Maria Henriksson, Emilia Wikstedt, Marie Mar- kelius, Lena Renner Real-time registration of listener reactions to unintelligibility in 127 misarticulated child speech Ivonne Contardo, Anita McAllister, Sofia Strömbergsson Session 8: Brain imaging and phonetics SUBIC - Stockholm University Brain Imaging Center and its signifi- 133 cance for humanistic and interdisciplinary research Francisco Lacerda, Björn Lindblom viii Proceedings from FONETIK 2014, Department of Linguistics, Stockholm University Voices after midnight – How a night out affects voice quality Alexandra Berger1,2, Rosanna Hedström Lindenhäll1,2, Mattias Heldner2, Sofia Karlsson1,2, Sarah Nyberg Pergament1,2, Ivan Vojnovic1,2, 1 Department of Clinical Science, Intervention and Technology, Division of Speech and Language Pathology, Karolinska Institutet, Sweden 2 Department of Linguistics, Stockholm University, Sweden [email protected], [email protected], [email protected], [email protected], [email protected], [email protected] Abstract This study aimed at examining how This study aimed to investigate how different acoustic voice quality parame- different parameters of the voice (jitter, ters were affected by the voice strain shimmer, LTAS and mean pitch) are induced by a night out talking in a noisy affected by a late night out. Three re- environment, and what effects can be cordings were made: one early evening observed the following day. The pa- before the night out, one after midnight, rameters examined in this study were and one on the next day. Each recording jitter (cycle-to-cycle variations in fre- consisted of a one minute reading and quency), shimmer (cycle-to-cycle varia- prolonged vowels. Five students took tions in amplitude) (Titze, 1995), LTAS part in the experiment. Results varied (long time average spectrum) and mean among the participants, but some pat- pitch. terns were noticeable in all parameters. Following Södersten, Ternström, & A trend towards increased mean pitch Bohman (2005) we expected the mean during the second recording was ob- pitch to increase and that LTAS would served among four of the subjects. indicate a decrease in vocal fry in the Somewhat unexpectedly, jitter and second recording. Furthermore, as pre- shimmer decreased between the first vious results imply that female speakers and second recordings and increased in tend to increase glottal closure after the third one. Due to the lack of ethical speaking in loud conditions (Linville, testing, only a small number of partici- 1995), we hypothesized that jitter and pants were included. A larger sample is shimmer would decrease continuously suggested for future research in order to from the first to the third recording. generalize results. Method Introduction To test our hypotheses we made three It is well known that the general vol- recordings and compared a number of ume at pubs, discotheques and similar voice quality measures in these. The venues is very loud and that as a guest first recording (R1) occurred at 7 pm on you have to raise your voice significant- a Friday evening. The subjects each ly in order to make yourself heard. read a text of approximately one minute Speakers tend to raise their voices in and then pronounced a prolonged [a]. loud conditions. This is known as the The second recording (R2) took place at Lombard effect (Lane & Tranel, 1971). half past midnight, after four hours in a This type of voice behavior can result bar, where background noise level was in vocal fatigue, temporary hoarseness measured. The third recording (R3) was and may in the long run cause vocal done at noon the next day. The subjects disorders (Vilkman, 2000). reported differences in sleep duration 1 Proceedings from FONETIK 2014, Department of Linguistics, Stockholm University (from 2 hours of sleep to 7 hours) as (R1, R2, R3) on three different voice well as differences in alcohol intake. quality measures: mean pitch, jitter, and shimmer. We used repeated contrasts to Equipment compare R1 vs. R2, and R2 vs. R3, The recordings were done in 16-bit, respectively. Mauchly’s test indicated 44.1 kHz with the application Røde rec that the assumption of spherizity was LE (version 2.8.1) for iPhone 4 (version met in all three ANOVAS, therefore we 7.0.3) and a Røde smartLav, tie clip,