National Institute of Standards and Technology における超伝導研究及び生活

https://www.neurones.espci.fr/Articles_PS/SPEECHCOM%201.pdf

Silent speech interfaces

Despite surveillance systems becoming increasingly ubiquitous in our living environment, automated surveillance, currently based on video sensory modality and machine intelligence, lacks most of the time the robustness and reliability required in several real applications. To tackle this issue, audio sensory devices have been incorporated, both alone or in combination with video, giving birth in the past decade, to a considerable amount of research. In this article, audio-based automated surveillance methods are organized into a comprehensive survey: A general taxonomy, inspired by the more widespread video surveillance field, is proposed to systematically describe the methods covering background subtraction, event classification, object tracking, and situation analysis. For each of these tasks, all the significant works are reviewed, detailing their pros and cons and the context for which they have been proposed. Moreover, a specific section is devoted to audio features, discussing their expressiveness and their employment in the above-described tasks. Differing from other surveys on audio processing and analysis, the present one is specifically targeted to automated surveillance, highlighting the target applications of each described method and providing the reader with a systematic and schematic view useful for retrieving the most suited algorithms for each specific requirement.

Audio Surveillance: A Systematic Review

Looking at the different points highlighted in this article, we affirm that forensic applications of speaker recognition should still be taken under a necessary need for caution. Disseminating this message remains one of the most important responsibilities of speaker recognition researchers.

/pdf/forensic-speaker-recognition-2hzwj7xbi2.pdf

Forensic speaker recognition

This paper presents the ALIZE/SpkDet open source software packages for text independent speaker recognition. This software is based on the well-known UBM/GMM approach. It includes also the latest speaker recognition developments such as Latent Factor Analysis (LFA) and unsupervised adaptation. Discriminant classifiers such as SVM supervectors are also provided , linked with the Nuisance Attribute Projection (NAP). The software performance is demonstrated within the framework of the NIST'06 SRE evaluation campaign. Several other applications like speaker diarization, embedded speaker recognition , password dependent speaker recognition and pathological voice assessment are also presented.

/pdf/alize-spkdet-a-state-of-the-art-open-source-software-for-1n5o7f776u.pdf

ALIZE/SpkDet: a state-of-the-art open source software for speaker recognition

This paper proposes an unsupervised method for real time detection of abnormal events in the context of audio surveillance. Based on training a One-Class Support Vector Machine (OC-SVM) to model the distribution of the normality (ambience), we propose to construct sets of decision functions. This modification allows controlling the trade-off between false-alarm and miss probabilities without modifying the trained OC-SVM that best capture the ambience boundaries, or its hyperparameters. Then we present an adaptive online scheme of temporal integration of the decision function output in order to increase performance and robustness. We also introduce a framework to generate databases based on real signals for the evaluation of audio surveillance systems. Finally, we present the performances obtained on the generated database.

/pdf/abnormal-events-detection-using-unsupervised-one-class-svm-2gak2bc4lg.pdf

Abnormal events detection using unsupervised One-Class SVM - Application to audio surveillance and evaluation -

The invention relates to a method for detecting abnormal audio events in a given environment, characterised in that said method comprises at least: an automated step of learning the modelling of the environment to be monitored, during which a database (9) is created by extracting acoustic parameters (2.1) associated with audio streams captured over a set period of time and unsupervised automatic segmentation (2.2) of said streams, followed by a step of grouping segments by classes (2.3) and statistical modelling of segment classes (2.4), an operational phase which includes analysing an audio stream, with extraction of acoustic parameters (3.1), a step of automatic segmentation (3.2) of said analysed stream substantially identical to that used during the learning phase and a step during which the likelihood (3.3) of every statistical model contained in the database is determined on each of the segments of the analysed audio stream, said step of determining likelihoods providing a likelihood value l which is compared with a threshold value in order to determine whether or not to trigger a signal indicating the presence or absence of anomalies in the analysed audio stream.

System and method for detecting abnormal audio events

In this paper we show that microphones that capture the bone-conducted voice can be used to improve Automatic Speech Recognition in noisy environments. These microphones exhibit good noise rejection properties and their signals are therefore good indications of speech activity, even in very noisy conditions. We conducted experiments where we used a throat microphone signal as a Voice Activity Detection (VAD) input signal and found that recognition accuracies in non-stationary noise improve significantly compared to when VAD is executed on a conventional microphone signal.

Improved speech recognition in noisy environments by using a throat microphone for accurate voicing detection

We propose a full reference visual quality metric to evaluate a semantic coding system which may not preserve exactly the position and/or the shape of objects. The metric is based on Scale-Invariant Feature Transform (SIFT) points. More specifically, Structural SIMilarity (SSIM) on windows around the SIFT points measures the compression artifacts (SSIM_SIFT). Conversely, the standard deviation of the matching distance between the SIFT points measures the geometric distortion (GEOMETRIC_SIFT). We validate our metric with subjective evaluation and reach a Spearman correlation of 0.86 for SSIM_SIFT and 0.74 for GEOMETRIC_SIFT.

https://hal.archives-ouvertes.fr/hal-00739928/document

A new object based quality metric based on SIFT and SSIM

Ad hoc networks are considered to provide flexible and robust communication in emergency scenarios like fire fighting. However, for the time being, it is not clear how existing digital wireless communication technologies perform in environments with fire, smoke and vapour. To investigate this issue, we carried out several experiments at the training facilities of the Paris fire brigade. Our goal was to evaluate the performance of standard wireless communication systems in a real fire fighting scenario. The main result is that wireless communication is not affected much by fire and smoke but is indeed affected by vapour. Technologies operating in the 2.4 GHz frequency achieve a higher communication range than those with 5 GHz in our setup.

/pdf/are-ieee-802-wireless-technologies-suited-for-fire-fighters-1vw8e7ko7i.pdf

Francois Capman

Papers

Abnormal events detection using unsupervised One-Class SVM - Application to audio surveillance and evaluation -

System and method for detecting abnormal audio events

Improved speech recognition in noisy environments by using a throat microphone for accurate voicing detection

A new object based quality metric based on SIFT and SSIM

Are IEEE 802 Wireless Technologies Suited for Fire Fighters