\input zb-basic
\input zb-ioport
\iteman{io-port 05351111}
\itemau{Szyma\'nski, Marcin; Grocholewski, Stefan}
\itemti{Error prediction-based semi-automatic segmentation of speech databases.}
\itemso{Sojka, Petr (ed.) et al., Text, speech and dialogue. 11th international conference, TSD 2008, Brno, Czech Republic, September 8--12, 2008. Proceedings. Berlin: Springer (ISBN 978-3-540-87390-7/pbk). Lecture Notes in Computer Science 5246. Lecture Notes in Artificial Intelligence, 543-550 (2008).}
\itemab
Summary: The manual segmentation of speech databases still outperforms the automatic segmentation algorithms and, at the same time, the quality of resulting synthetic voice depends on the accuracy of the phonetic segmentation. In this paper we describe a semi-automatic speech segmentation procedure, in which a human expert manually allocates the selected boundaries prior to the automatic segmentation of the rest of the corpus. Segmentation error predictor is designed, estimated and then used to generate a sequence of manual annotations done by an expert. The obtained error response curves are significantly better than random segmentation strategies. The results are presented for two different Polish corpora.
\itemrv{~}
\itemcc{}
\itemut{}
\itemli{doi:10.1007/978-3-540-87391-4\_69}
\end