Towards Semantic Analysis of Conversations: A System for the Live Identification of Speakers in Meetings
2008 IEEE International Conference on Semantic Computing (2008)
Aug. 4, 2008 to Aug. 7, 2008
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICSC.2008.58
In the following article we present an application that enables online identiﬁcation of who is currently speaking using a single far-ﬁeld microphone in a meeting scenario. By leveraging techniques from both the ﬁeld of speaker identiﬁcation and speaker diarization, the system is able to recognize the current speaker after any two seconds of speech. An evaluation of the robustness of the algorithm using the AMI Meeting Corpus and the NIST Speaker Di- arization Development set resulted in a Diarization Error Rate of 12.67%.
speaker, diarization, identification, meetings, online, realtime
G. Friedland and O. Vinyals, "Towards Semantic Analysis of Conversations: A System for the Live Identification of Speakers in Meetings," 2008 IEEE International Conference on Semantic Computing(ICSC), vol. 00, no. , pp. 426-431, 2008.