Functional data analysis of tongue articulation in Gothenburg and Malmöhus Swedish /i:, y:, u-:/

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingResearchpeer-review

136 Downloads (Pure)


Articulatory data collected from nine speakers each of Gothenburg and Malmöhus Swedish were used in a Functional Data Analysis (FDA) to study tongue articulation dynamics, more specifically the height and frontness of the tongue body and tip in the palatal vowels /i ̆, y ̆, u̟ ̆/. Standard z-score transformations were used for speaker normalisation. Results showed that the tongue articulation for /i ̆/ and /y ̆/ is generally similar, and significantly different from /u̟ ̆/ in both Malmöhus and Gothenburg Swedish. We also found a subdivision of Gothenburg Swedish into two subtypes, where type 1 resembled Malmöhus Swedish more. Significant differences in tongue body height were found between all varieties for all of the vowels, except for /y ̆/ between Gothenburg type 1 and Malmöhus Swedish.
Original languageEnglish
Title of host publication[Host publication title missing]
EditorsRobert Eklund
PublisherLinköping University
Number of pages4
Publication statusPublished - 2013
EventFonetik 2013 - Linköping, Linköping, Sweden
Duration: 2013 Jun 122013 Jun 13


ConferenceFonetik 2013

Subject classification (UKÄ)

  • General Language Studies and Linguistics


  • Swedish vowels
  • Functional data analysis
  • Articulography


Dive into the research topics of 'Functional data analysis of tongue articulation in Gothenburg and Malmöhus Swedish /i:, y:, u-:/'. Together they form a unique fingerprint.

Cite this