We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Extracting Relationship of Meeting Minutes Generated by Speech Recognition System using Entity Resolution.
- Authors
Seikoh Nishita; Motoki Itoh
- Abstract
A minutes generation system by speech recognition automatically records minutes generated from voice data in the meeting. If generated minutes were not strictly managed, the minutes would possibly include error words caused by the speech recognition. Such error words makes information retrieval on minutes difficult. To address the problem, this paper proposes a technique to extract relationship of minutes generated by the speech recognition. Our technique is based on "collective entity resolution in relational data." This paper also reports an experimental evaluation of our technique. The experimental result suggests effectiveness of the technique for minutes texts including error words.
- Subjects
AUTOMATIC speech recognition; TEXT mining; MEETING minutes; PROBLEM solving; INFORMATION retrieval
- Publication
IAENG International Journal of Computer Science, 2016, Vol 43, Issue 3, p1
- ISSN
1819-656X
- Publication type
Article