20.500.12556/RUL-190073 Annotating disfluency in speech corpora Označevanje netekočnosti v govornih korpusih The paper investigates disfluencies in speech, defined as phenomena that typically disrupt the flow of speech and do not contribute to its primary propositional content. These phenomena typically include filled and silent pauses, self-repairs, repetitions, and related features. Rather than being treated as errors, disfluencies are conceptualized as an integral component of spoken interaction. A review of the most influential annotation schemes reveals that no single framework has been established as a generally accepted standard. Combining an inductive, data-driven approach with a critical analysis of existing categorizations, the study proposes a tailored, multi-layer annotation scheme. This scheme covers a broad range of phenomena previously identified as disfluencies in earlier frameworks. The evaluation demonstrates that the scheme is sufficiently clear and unambiguous, offers well-defined and appropriate categories, and is comparable to other contemporary schemes. The study contributes to the development of spoken-language resources for Slovenian and provides a foundation for future linguistic, psycholinguistic, and technological research. Prispevek obravnava netekočnosti (angl. disfluencies) v govoru, to so pojavi, ki prekinjajo tok govora in ne prispevajo nujno k propozicijski vsebini. Tipično vključujejo zapolnjene in tihe premore, samopopravljanja, ponavljanja ipd. Netekočnosti ne obravnavamo kot napake, ampak kot sestavni del govorne interakcije. Po pregledu najvplivnejših označevalnih shem ugotovimo, da se nobena ni uveljavila kot splošno sprejeti standard. S kombinacijo induktivnega pristopa, izhajajočega iz empiričnih podatkov, in kritičnega pregleda obstoječih kategorizacij razvijemo prilagojeno večplastno označevalno shemo. Ta vključuje širok nabor pojavov, ki so bili prepoznani kot netekočnosti v predhodnih shemah. Evalvacija kaže, da je shema primerno jasna in enoumna, da ima definirane ustrezne kategorije in je primerljiva z drugimi sodobnimi shemami. Raziskava prispeva k razvoju govornih virov za slovenščino ter nudi osnovo za nadaljnje jezikoslovne, psiholingvistične in tehnološke študije. filled pause repetition self-repair spoken language prosody zapolnjeni govor ponavljanje samopopravljanje govorjeni jezik prozodija true false true Angleški jezik Slovenski jezik Članek v reviji 2026-10-21 16:52:59 2026-10-21 16:53:06 2026-10-22 04:13:21 0000-00-00 00:00:00 2026 0 0 Str. 135-168 št. 1 Letn. 14 2026 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 2026-10-09 811.163.6 2335-2736 10.4312/slo2.0.2026.1.135-168 294614275 Verdonik.pdf Verdonik.pdf 1 E3F535610A1DC2EBEE443A6117F4460B 18904a559334927543273f2866d1a9de23b42e25216352e6f19578af70fb0c8d 187b2336-cd5f-11f1-8bc5-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=251095 https://journals.uni-lj.si/slovenscina2/article/view/22594 1 ed8d2857-cd5e-11f1-8bc5-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=251094 Filozofska fakulteta Slovenščina 2.0 0 0 0