Towards Audio-Visual On-line Diarization Of Participants In Group Meetings
Hayley Hung, Gerald H Friedland · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2008
Abstract. We propose a fully automated, unsupervised, and non-intrusive method of identifying the current speaker audio-visually in a group conversation. This is achieved without specialized hardware, user interaction, or prior assignment of microphones to participants. Speakers are identified acoustically using a novel on-line speaker diarization approach. The output is then used to find the corresponding person in a four-camera video stream by approximating individual activity with computationally efficient features. We present results showing the robustness of the association on over 4.5 hours of non-scripted audio-visual meeting data. 1