Segmental Duration in Utterance-Initial Environment: Evidence from Finnish Speech Corpora

  • Tuomo Saarni
  • Jussi Hakokari
  • Jouni Isoaho
  • Olli Aaltonen
  • Tapio Salakoski
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4139)


This study examines segmental durations produced by Finnish speakers in utterance-initial environments. We have established a method to statistically examine segmental duration on the phone level in speech corpora. The two corpora represented in this study consist mainly of television news broadcasts and short texts read aloud by professional speakers. Previous studies conducted have been contradictory; there are reports of initial shortening in certain languages and lengthening in others. Our results are conclusive in neither way, but suggest a qualitatively differentiated behavior. We have observed lengthening of all utterance-initial vowels, diphthongs included, and shortening of phonologically long plosive (stop) consonants. No other speech sounds are significantly affected. These findings hold in both corpora, in despite of different speakers and annotators.


Speech Sound Speech Synthesis Speech Corpus Short Vowel Stressed Syllable 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Campbell, N.: Segmental Elasticity and Timing in Japanese Speech. In: Tohkura, Y., Vatikiotis-Bateson, E., Sagisaka, Y. (eds.) Speech Perception, Production, and Linguistic Structure, pp. 403–418. IOS Press, Amsterdam (1992)Google Scholar
  2. 2.
    Chung, H., Gim, G., Huckvale, M.: Consonantal and Prosodic Influences on Korean Vowel Duration. In: Proceedings of Eurospeech, Budapest, Hungary, vol. 2, pp. 707–710 (1999)Google Scholar
  3. 3.
    Duez, D.: Acoustic Correlates of Subjective Pauses. Journal of Psycholinguistic Research 22(1), 21–39 (1993)CrossRefGoogle Scholar
  4. 4.
    Greenberg, J.: Language Universals: with Special Reference to Feature Hierarchies. Mouton (1966)Google Scholar
  5. 5.
    Hakokari, J., Saarni, T., Salakoski, T., Isoaho, J., Aaltonen, O.: Determining Prepausal Lengthening for Finnish Rule-Based Speech Synthesis. In: Proceedings of Speech Analysis, Synthesis and Recognition: Applications of Phonetics (SASR 2005), Krakow, Poland (2005)Google Scholar
  6. 6.
    Hansson, P.: Prosodic Phrasing in Spontaneous Swedish. Academic Dissertation. Travaux de l’institut de linguistique de Lund, vol. 43. Lund University, Lund (2003)Google Scholar
  7. 7.
    Hockey, B.A., Fagyal, Z.: Phonemic Length and Pre-Boundary Lengthening: an Experimental Investigation on the Use of Durational Cues in Hungarian. In: Proceedings of the XIVth International Congress of Phonetics Sciences, San Francisco, pp. 313–316 (1999)Google Scholar
  8. 8.
    Kaiki, N., Takeda, K., Sakisaga, Y.: Statistical Analysis for Segmental Duration Rules in Japanese Speech Synthesis. In: Proceedings of the 1990 International Conference on Spoken Language Processing, Kobe, Japan, pp. 17–20 (1990)Google Scholar
  9. 9.
    Krull, D.: Prepausal Lengthening in Estonian: Evidence from Conversational Speech. In: Lehiste, I., Ross, J. (eds.) Estonian Prosody: Papers from a Symposium, Proceedings of the International Symposium on Estonian Prosody, Tallinn, Estonia, pp. 136–148. Institute of Estonian Language, Tallinn (1997)Google Scholar
  10. 10.
    Nagano-Madsen, Y.: Temporal Characteristics in Eskimo and Yoruba: a Typological Consideration. In: The Sixth Swedish Phonetics Conference, Göteborg (1992)Google Scholar
  11. 11.
    Vainio, M.: Artificial Neural Network Based Prosody Models for Finnish Text-to-Speech Synthesis. Academic dissertation, University of Helsinki (2001)Google Scholar
  12. 12.
    White, L.S.: English Speech Timing: a Domain and Locus Approach. University of Edinburgh PhD dissertation (2002) Google Scholar
  13. 13.
    Zu, Y., Chen, X.: Segmental Durations of a Labelled Speech Database and its Relation to Prosodic Boundaries. In: Proceedings of the 1st International Symposium on Chinese Spoken Language Processing (ISCSLP 1998) (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Tuomo Saarni
    • 1
  • Jussi Hakokari
    • 2
  • Jouni Isoaho
    • 1
  • Olli Aaltonen
    • 2
  • Tapio Salakoski
    • 1
  1. 1.Turku Centre for Computer ScienceFinland
  2. 2.Phonetics LaboratoryUniversity of TurkuFinland

Personalised recommendations