
Abstract
The relation between symbolic and signal features of prosodic boundaries is experimentally studied using prediction methods. Text-based break index prediction turns out to be fairly good, but signal-based prediction and pause duration prediction perform worse. A possible reason is that random signal feature variations, as usually produced by humans, are hard to predict.
Dokumententyp: | Konferenzbeitrag (Paper) |
---|---|
Fakultät: | Sprach- und Literaturwissenschaften > Department 2 > Phonetik und Sprachverarbeitung |
Themengebiete: | 400 Sprache > 400 Sprache |
URN: | urn:nbn:de:bvb:19-epub-13159-6 |
Sprache: | Englisch |
Dokumenten ID: | 13159 |
Datum der Veröffentlichung auf Open Access LMU: | 06. Jul. 2012, 11:13 |
Letzte Änderungen: | 04. Nov. 2020, 12:53 |