A speech-based and binaural Speech Transmission Index is presented and evaluated in a variety of acoustical degradations and spatial conditions. The proposed method facilitates the assessment of speech intelligibility in classical room acoustics and electroacoustics by simply comparing a binaural speech recording in adverse conditions with its clean original. Both the binaural processing stage and the speech-based Speech Trans- mission Index method are e ective and computationally fast realizations. The central part of the binaural processor forms a cross-correlation stage that is designed to replicate psychoacoustic data of binaural in- teraction. Supplemented with the head shadow e ect, which is generated in a `better-ear' fashion, a fair amount of the binaural advantage in speech intelligibility is modeled. An evaluation of the method was performed in a battery of listening tests. These tests incorporated di erent degradations, as e.g., stationary noise and uctuation noise, a set of nonlinear signal alterations, including a speech enhancement processor, and a multitude of spatial con gurations with di erent room acoustics and with up to four interferers. As a result, the objective method o ers a stable prediction of the subjective results in binaural speech intelligibility throughout most of the linear degradations. In spite of that, the full account of the binaural advantage is not achieved by the current implementation of the method, which suggests further research.
Authors:
Schlesinger, Anton; Ramirez, Juan-Pablo; Boone, Marinus M.
Affiliations:
Deutsche Telekom Laboratories, Technische Universität Berlin, Berlin, Germany; Delft University of Technology, Delft, The Netherlands(See document for exact affiliation information.)
AES Conference:
40th International Conference: Spatial Audio: Sense the Sound of Space (October 2010)
Paper Number:
4-6
Publication Date:
October 8, 2010
Subject:
Perception and Evaluation of Spatial Audio
Click to purchase paper as a non-member or you can login as an AES member to see more options.
No AES members have commented on this paper yet.
To be notified of new comments on this paper you can subscribe to this RSS feed. Forum users should login to see additional options.
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.