Paper: JURISDIC: Polish Speech Database for Taking Dictation of Legal Texts

ACL ID L08-1499
Title JURISDIC: Polish Speech Database for Taking Dictation of Legal Texts
Venue International Conference on Language Resources and Evaluation
Session Main Conference
Year 2008
Authors

y, two broad categories can be observed. A class of systems, like those inspired by (Grefenstette, 1999), (Silberztein, 2000), (Boguraev, 2000), (Grover et al., 2000), (Simov et al., 2002), essentially deconstruct an annotations store data structure into a string (suitably adorned with in-line annotation boundary information) and apply FS m range of named entity detection systems, in a variety of domains. Named entity detection has typically been interleaved with shallow syntactic parsing, also implemented as a cascade of AFst grammars (Boguraev, 2000). The ability to mix, within the same application, syntactic and semantic operations over an annotations store offers not just well known benefits like generalizing over syntactic configurations with <...