A real-time speech enhancement framework for multi-party meetings

Abstract

This paper proposes a real-time speech enhancement framework working in presence of multiple sources in reverberated environments. The aim is to automatically reduce the distortions introduced by room reverberation in the available distant speech signals and thus to achieve a significant improvement of speech quality for each speaker. The overall framework is composed by three cooperating blocks, each one fulfilling a specific task: speaker diarization, room-impulse response identification and speech dereverberation. In particular the speaker diarization algorithm is essential to pilot the operations performed in the other two stages in accordance with speakers' activity in the room. Extensive computer simulations have been performed by using a subset of the AMI database: Obtained results show the effectiveness of the approach. © 2011 Springer-Verlag.