Patricia Scanlon

Image transforms, such as the discrete cosine, are widely used to extract visual features from the speaker's mouth region to be used in automatic speechreading and audio-visual speech recog- nition. Typically, the spatial frequency components with the highest energy in the transform space are retained for recogni- tion. This paper proposes an alternative technique for select- ing such features, by

ABSTRACT While automated condition monitoring of rotating machines often use vibration signals for defect detection, diagnosis, and residual life predictions, in this paper, the acoustic noise signal (<; 25 kHz), acquired via non-contact microphone sensors, is used to predict the remaining useful life (RUL). Modulation spectral (MS) analysis of acoustic signals has the potential to provide additional long-term information over more conventional short-term signal spectral components. However, the high dimensionality of MS features has been cited as a limitation to their applicability in this area in the literature. Therefore, in this study, a novel approach is proposed which employs an information theoretic approach to feature subset selection of modulation spectra features. This approach does not require information regarding the spectral location of defect frequencies to be known or pre-estimated and leverages information regarding the chronological order of data samples for dimensionality reduction. The results of this study show significant improvements for this proposed approach over the other commonly used spectral-based approaches for the task of predicting RUL by up to 19% relative over the standard envelope analysis approach used in the literature. A further 16% improvement was achieved by applying a more rigorous approach to labeling of acoustic samples acquired over the lifetime of the machines over a fixed length class labeling approach. A detailed misclassification analysis is provided to interpret the relative cost of system errors for the task of residual life predictions of rotating machines used in industrial applications.

... [1]. It is well known that visual information from the face of a speaker provides related speech ... time, good speech recognition performance could be obtained using the raw data signal as input to a ... perform better than using high-level geometric measurements as visual features. ...

It has been found that auditory-visual speech perception differs over languages, particularly between English and Japanese speakers. This difference emerges due to increased use of visual information by English speakers between 6 and 8 years [14]. This study investigates the linguistic factors that may cause changes in auditory-visual speech perception development. Children aged between 5 and 8 years were given tests of reading, articulation, language-specific speech perception and auditory-visual speech perception. The results ...

ABSTRACT Radio frequency (RF) fingerprinting is a technique which attempts to extract a unique identifier from wireless signal transmissions in order to perform automated device identification by exploiting variations in the transmitted signal caused by hardware and manufacturing inconsistencies. The problem of signaling storms caused by increased core network signaling load due to idle mode cell camping on femtocells is addressed and a novel approach to RF fingerprinting is proposed as a solution to tackle this problem by identifying device model types. This study describes a large data set containing 54 Universal Mobile Telecommunications System (UMTS) user equipment (UE) devices (41 model types), the largest of its kind reported in the literature. This study also presents a novel feature extraction technique, which greatly improves identification accuracy over standard spectral approaches while limiting the number of random access channel (RACH) preambles required for identification. Accuracy of 99.8 percent was achieved. Importantly, the proposed technique can be implemented using today's low cost high‐volume receivers and requires no manual performance tuning. © 2010 Alcatel‐Lucent. © 2010 Wiley Periodicals, Inc.

The ability to predict the Remaining Useful Life (RUL) of Rotating Machines is a highly desirable function of Automated Condition Monitoring (ACM) systems. Typically, vibration signals are acquired through contact with the machine and used for monitoring. In this paper, a novel implementation of the ubiquitous feature extraction approach Envelope Analysis (EA) is applied to acoustic noise signals (< 25