An artificial neural network approach for predicting architectural speech security (L)

Abstract

/npsi/ctrl?lang=en http://nparc.cisti-icist.nrc-cnrc.gc.ca/npsi/ctrl?lang=fr Access and use of this website and the material on it are subject to the Terms and Conditions set forth at http://nparc.cisti-icist.nrc-cnrc.gc.ca/npsi/jsp/nparc_cp.jsp?lang=en NRC Publications Archive Archives des publications du CNRCThis publication could be one of several versions: author's original, accepted manuscript or the publisher's version. / La version de cette publication peut être l'une des suivantes : la version prépublication de l'auteur, la version acceptée du manuscrit ou la version de l'éditeur. America, 117, 4, pp. 1709America, 117, 4, pp. -1712America, 117, 4, pp. , 2005 An Artificial neural network approach for predicting architectural speech security (L) Xu, J.; Bradley, J. S.; Gover, B. N. Journal of the Acoustical Society of ABSTRACTSignal-to-noise (S/N) type measures have been developed for predicting architectural speech privacy and speech security, which is required to accurately rate the probability of a listener outside a room being able to overhear conversations from within the room. However, these measures may not be ideal for speech security situations. In the present work, an approach that uses the artificial neural networks to directly represent the functional relationship between the octave band (250 Hz -8 kHz) S/N ratios and the speech intelligibility score and security thresholds has been investigated. The artificial neural network (ANN) approach provides a direct and accurate method for predicting the speech intelligibility score and security thresholds.

Neural Oscillations in Speech: Don't be Enslaved by the Envelope

“…However, the speech envelope is readily obscured in noisy backgrounds and reverberant environments (Houtgast and Steeneken, 1985) and intact spectral content can be used by the listener to at least partially compensate for degraded temporal envelope information (Sheft et al, 2008). Indeed, although the temporal envelope of speech has been shown to be very important for comprehension (e.g., Drullman et al, 1994a,b) there is good evidence that the spectral content of the speech signal is at least as decisive for speech intelligibility (if not more so; Xu et al, 2005; Lorenzi et al, 2006; Luo and Poeppel, 2007; Obleser et al, 2008; Obleser and Weisz, 2012; Scott and Mcgettigan, 2012). Moreover, it has recently been suggested that the temporal envelope and spectral content of natural speech (or conspecific vocalizations in non-human animals) are non-independent, and that speech comprehension performance is in fact best predicted from the presence of a “core” spectrotemporal modulation region in the modulation transfer function of a stimulus (Elliott and Theunissen, 2009).…”

Predicting Speech Security and Speech Privacy Casual Versus Intentional Listening

“…These differences mean that although speech privacy measures are sometimes used to rate the speech security of a structure, this may not be appropriate, or even possible in some cases. Other speech security researchers have also identified the lack of applicability of speech privacy-based measures to speech security prediction [2] . It should be recognised that small errors in the prediction…”

How private is your consultation? Acoustic and audiological measures of speech privacy in the otolaryngology clinic

“…One aspect of privacy that is often overlooked is the right to conduct a consultation without being overheard. Following the passing of the HIPPA act a number of studies have attempted to quantify the privacy of healthcare environments, commonly referred to as ''architectural speech privacy'' [7,10,15,17]. These studies use acoustic measurements of sound transmission between two points, the most widely used measurement being the articulation index (AI) [8,12,16,17].…”