A self-starting multistage, time-domain procedure is presented for the identification of nonlinear, multi-degree-of-freedom systems undergoing free oscillations or subjected to arbitrary direct force excitations and/or nonuniform support motions. Recursive least-squares parameter estimation methods combined with nonparametric identification techniques are used to represent, with sufficient accu...

This article describes the systems jointly submitted by Institute for Infocomm (IR), the Laboratoire d’Informatique de l’Universit du Maine (LIUM), Nanyang Technology University (NTU) and the University of Eastern Finland (UEF) for 2015 NIST Language Recognition Evaluation (LRE). The submitted system is a fusion of nine sub-systems based on i-vectors [1] extracted from different types of featur...

In this study, we explore a human benchmark in language recognition, for the purpose of comparing human performance to machine performance in the context of the NIST LRE 2007. Humans are categorised in terms of language proficiency, and performance is presented per proficiency. The main challenge in this work is the design of a test and application of a performance metric which allows a meaning...

In this paper, we introduce a dialect recognition method that makes use of phonetic models adapted per dialect without phonetically labeled data. We show that this method can be implemented efficiently within an existing PRLM[1] system. We compare the performance of this system with other state-of-theart dialect recognition methods (both acoustic and token-based) on the NIST LRE 2007 English an...

We introduce a unified framework for nonlinear signal processing with finite normal mixtures (FNM) by using maximum partial likelihood (MPL) theory. We show that the equivalence of MPL to accumulated relative entropy (ARE) minimization is valid for the FNM. Then, we define the information geometry of MPL and use the result to derive the em algorithm for distribution learning based on the FNM mo...

The photocatalytic activity for CO(2) reduction and optical and electrochemical properties of two Ru(II)-Re(I) binuclear complexes [Ru(dmb)(2)LRe(CO)(3)Cl](2+) (Ru-Re and Ru=Re, dmb = 4,4'-dimethyl-2,2'-bipyridine) with 1,2-bis(4'-methyl-2,2'-bipyridyl-4-yl)ethane and 1,2-bis(4'-methyl-2,2'-bipyridyl-4-yl)ethene as bridging ligands have been investigated. The conjugation content of the bridging...

We have I,een investigating a novel integratioil of sensing aiitl coiiipressioii on an image sensor. By the int egrat ion. nllinl,er of pixels in the image signal that lias to 1,e r e a d o ~ ~ t from the sensor can he significantly retl~icetl. ant1 the integration can conse-cl~leiit ly illcrease the pixel rate of t lie sensor. In this paper. we present a new compression sensor which lias 128 x...

Spoken language recognition requires a series of signal processing steps and learning algorithms to model distinguishing characteristics of different languages. In this paper, we present a sparse discriminative feature learning framework for language recognition. We use sparse coding, an unsupervised method, to compute efficient representations for spectral features from a speech utterance whil...

This paper discusses methodological issues related to natural language modeling in the framework of the LRE project ANTHEM 1. The objective of ANTHEM is to develop a portable prototype of a multilingual natural language interface that allows users of Healthcare Information Systems to enter diagnostic expressions using their own natural language, and to have this input translated in whatever for...

Gaussian mixture model support vector machine (GMMSVM) with nuisance attribute projection (NAP) has been found to be effective and reliable for speaker and language recognition. In maximum a posteriori (MAP) adaptation of GMM, the relevance factor is the parameter that regulates how much the adaptation data affect the base model, which impacts the final recognition performance. In our previous ...