ISCA Archive Interspeech 2021
ISCA Archive Interspeech 2021

Glottal Stops in Upper Sorbian: A Data-Driven Approach

Ivan Kraljevski, Maria Paola Bissiri, Frank Duckhorn, Constanze Tschoepe, Matthias Wolff

We present a data-driven approach for the quantitative analysis of glottal stops before word-initial vowels in Upper Sorbian, a West Slavic minority language spoken in Germany. Glottal stops are word-boundary markers and their detection can improve the performance of automatic speech recognition and speech synthesis systems.

We employed cross-language transfer using an acoustic model in German to develop a forced-alignment method for the phonetic segmentation of a read-speech corpus in Upper Sorbian. The missing phonemic units were created by combining the existing phoneme models. In the forced-alignment procedure, the glottal stops were considered optional in front of word-initial vowels.

To investigate the influence of speaker type (males, females, and children) and vowel on the occurrence of glottal stops, binomial regression analysis with a generalized linear mixed model was performed. Results show that children glottalize word-initial vowels more frequently than adults, and that glottal stop occurrences are influenced by vowel quality.


doi: 10.21437/Interspeech.2021-1101

Cite as: Kraljevski, I., Bissiri, M.P., Duckhorn, F., Tschoepe, C., Wolff, M. (2021) Glottal Stops in Upper Sorbian: A Data-Driven Approach. Proc. Interspeech 2021, 1001-1005, doi: 10.21437/Interspeech.2021-1101

@inproceedings{kraljevski21_interspeech,
  author={Ivan Kraljevski and Maria Paola Bissiri and Frank Duckhorn and Constanze Tschoepe and Matthias Wolff},
  title={{Glottal Stops in Upper Sorbian: A Data-Driven Approach}},
  year=2021,
  booktitle={Proc. Interspeech 2021},
  pages={1001--1005},
  doi={10.21437/Interspeech.2021-1101}
}