There is extensive research on listening-room compensation algorithms to remove the unwanted degradations of speech quality. Objective measures for quality assessment are required in order to evaluate such algorithms. They exist in two classes: reverberation-cancellation algorithms that equalize the room impulse response of the acoustic channel, and reverberation-suppression algorithms that remove the reverberant part of the speech signal by calculating the spectral weight for time-frequency coefficients. This paper focuses on evaluating the former. Algorithms are analyzed regarding their capability to assess reverberation, coloration, spectral distortion, perceived distance, and overall quality of the signals. An evaluation of the sound quality of the dereverberated signals is derived from subjective listening tests and then compared to objective measures.
Goetze, Stefan; Albertin, Eugen; Rennies, Jan; Habets, Emanuël A.P.; Kammeyer, Karl-Dirk
Affiliations: Fraunhofer Institute for Digital Media Technology, IDMT, Hearing, Speech and Audio Technology, Oldenburg, Germany; International Audio Laboratories Erlangen, University of Erlangen-Nuremberg, Erlangen, Germany; University of Bremen, Dept. of Communications Engineering, Bremen, Germany(See document for exact affiliation information.)
JAES Volume 62 Issue 6 pp. 386-399; June 2014
Publication Date: July 2, 2014
No AES members have commented on this paper yet.
If you are not yet an AES member and have something important to say about this paper then we urge you to join the AES today and make your voice heard. You can join online today by clicking here.