Javier Tejedor

Toward Prevention of Pipeline Integrity Threats Using a Smart Fiber-Optic Surveillance System

This paper presents the first available report in the literature of a system aimed at the detection and classification of threats in the vicinity of a long gas pipeline. The system is based on phase-sensitive optical time-domain reflectometry technology for signal acquisition and pattern recognition strategies for threat identification. The system operates in two different modes: 1) machine+activity identification, which outputs the activity being carried out by a certain machine; and 2) threat detection, aimed at detecting threats no matter what the real activity being conducted is. Different strategies dealing with position selection and normalization methods are presented and evaluated using a rigorous experimental procedure on realistic field data. Experiments are conducted with eight machine+activity pairs, which are further labeled as threat or nonthreat for the second mode of the system. The results obtained are promising given the complexity of the task and open the path to future improvements toward fully functional pipeline threat detection systems operating in real conditions.

A Novel Fiber Optic Based Surveillance System for Prevention of Pipeline Integrity Threats

This paper presents a novel surveillance system aimed at the detection and classification of threats in the vicinity of a long gas pipeline. The sensing system is based on phase-sensitive optical time domain reflectometry (ϕ-OTDR) technology for signal acquisition and pattern recognition strategies for threat identification. The proposal incorporates contextual information at the feature level and applies a system combination strategy for pattern classification. The contextual information at the feature level is based on the tandem approach (using feature representations produced by discriminatively-trained multi-layer perceptrons) by employing feature vectors that spread different temporal contexts. The system combination strategy is based on a posterior combination of likelihoods computed from different pattern classification processes. The system operates in two different modes: (1) machine + activity identification, which recognizes the activity being carried out by a certain machine, and (2) threat detection, aimed at detecting threats no matter what the real activity being conducted is. In comparison with a previous system based on the same rigorous experimental setup, the results show that the system combination from the contextual feature information improves the results for each individual class in both operational modes, as well as the overall classification accuracy, with statistically-significant improvements.

Deep multi-view representation learning for multi-modal features of the schizophrenia and schizo-affective disorder

This work is originated from the MLSP 2014 Classification Challenge which tries to automatically detect subjects with schizophrenia and schizo-affective disorder by analyzing multi-modal features derived from magnetic resonance imaging (MRI) data. We employ Deep Neural Network (DNN)-based multi-view representation learning for combining multimodal features. The DNN-based multi-view models include deep canonical correlation analysis (DCCA) and deep canonically correlated auto-encoders (DCCAE). In addition, support vector machine with Gaussian kernel is used to conduct classification with the compact bottleneck features learned by the deep multi-view models. Our experiments on the dataset provided by the MLSP Classification Challenge show that bottleneck features learned via deep multi-view models obtain better results than the trimming features used in the baseline system in terms of the receiver operating characteristic (ROC) area under the curve (AUC).

Early detection of pipeline integrity threats using a smart fiber optic surveillance system: the PIT-STOP project

The preliminary results of a surveillance system set up for real time monitoring activities along a pipeline and analyzing for possible threats are presented. The system consists of a phi-OTDR based sensor used to monitor vibrations along an optical fiber combined with a pattern recognition system that classifies the recorded signals. The acoustic traces generated by the activities of different machines at various locations along a pipeline were recorded in the field. The signals, corresponding to machinery activities, were clearly distinguished from background noise. A threat classification rate of 68.11% with 55.55% false alarms was obtained.

Emotional adaptive training for speaker verification

Speaker verification suffers from significant performance degradation with emotion variation. In a previous study, we have demonstrated that an adaptation approach based on MLLR/CMLLR can provide a significant performance improvement for verification on emotional speech. This paper follows this direction and presents an emotional adaptive training (EAT) approach. This approach iteratively estimates the emotion-dependent CMLLR transformations and re-trains the speaker models with the transformed speech, which therefore can make use of emotional enrollment speech to train a stronger speaker model. This is similar to the speaker adaptive training (SAT) in speech recognition. The experiments are conducted on an emotional speech database which involves speech recordings of 30 speakers in 5 emotions. The results demonstrate that the EAT approach provides significant performance improvements over the baseline system where the neutral enrollment data are used to train the speaker models and the emotional test utterances are verified directly. The EAT also outperforms another two emotionadaptation approaches in a significant way: (1) the CMLLR-based approach where the speaker models are trained with the neutral enrollment speech and the emotional test utterances are transformed by CMLLR in verification; (2) the MAP-based approach where the emotional enrollment data are used to train emotion-dependent speaker models and the emotional utterances are verified based on the emotion-matched models.

Robust submodular data partitioning for distributed speech recognition

Distributed deep neural networks are commonly employed for building automatic speech recognition (ASR) systems. In this work, we employ the robust submodular partitioning approach, which aims to split the training data into small disjoint data subsets and use each of these subsets to train a particular deep neural network. Two efficient algorithms are used as robust submodular functions [1], namely `Greedi-Max and `Minorization-Maximization [2], which are guaranteed to provide tight approximations to the submodular data partition problem. Experiments on TIMIT database show that each of the distributed neural networks trained by the submodular data subset obtains better results than that trained on any subset of data partitioned in a random way., In addition, multi-class adaboost is effectively used to fuse the outputs of the deep neural networks and provides competitive ASR results compared with the traditional ASR system. Besides, the time incurred by acoustic modeling is significantly reduced, which delivers us further benefits.

Similar Word Model for Unfrequent Word Enhancement in Speech Recognition

The popular n-gram language model (LM) is weak for unfrequent words. Conventional approaches such as class-based LMs pre-define some sharing structures (e.g., word classes) to solve the problem. However, defining such structures requires prior knowledge, and the context sharing based on these structures is generally inaccurate. This paper presents a novel similar word model to enhance unfrequent words. In principle, we enrich the context of an unfrequent word by borrowing context information from some “similar words.” Compared to conventional class-based methods, this new approach offers a fine-grained context sharing by referring to words that best match the target word, and it is more flexible as no sharing structures need to be defined by hand. Experiments on a large-scale Chinese speech recognition task demonstrated that the similar word approach can improve performance on unfrequent words significantly, while keeping the performance on general tasks almost unchanged.

Towards detection of pipeline integrity threats using a SmarT fiber-OPtic surveillance system: PIT-STOP project blind field test results

This paper presents the first report on on-line and final blind field test results of a pipeline integrity threat surveillance system. The system integrates a machine+activity identification mode, and a threat detection mode. Two different pipeline sections were selected for the blind tests: One close to the sensor position, and the other 35 km away from it. Results of the machine+activity identification mode showed that about 46% of the times the machine, the activity or both were correctly identified. For the threat detection mode, 8 out of 10 threats were correctly detected, with 1 false alarm.

ATVS-CSLT-HCTLab System for NIST 2013 Open Keyword Search Evaluation

This paper presents the ATVS-CSLT-HCTLab spoken term detection STD system submitted to the NIST 2013 Open Keyword Search evaluation. The evaluation consists of searching a list of query terms in Vietnamese conversational speech data. Our submission involves an automatic speech recognition ASR subsystem which converts speech signals into word/phone lattices, and an STD subsystem which indexes and searches for query terms. The submission is a hybrid approach which employs a word-based system to search for in-vocabulary INV terms and a phone-based system to search for out-of-vocabulary OOV terms. A term-dependent discriminative confidence estimation is employed to score confidence of detections. Although the ASR performance is not state-of-the-art, our submission achieves a moderate STD performance in the evaluation.