Speaker attribution of multiple telephone conversations using a complete-linkage clustering approach

In this paper we propose and evaluate a speaker attribution system using a complete-linkage clustering method. Speaker attribution refers to the annotation of a collection of spoken audio based on speaker identities. This can be achieved using diarization and speaker linking. The main challenge associated with attribution is achieving computational efficiency when dealing with large audio archives. Traditional agglomerative clustering methods with model merging and retraining are not feasible for this purpose. This has motivated the use of linkage clustering methods without retraining. We first propose a diarization system using complete-linkage clustering and show that it outperforms traditional agglomerative and single-linkage clustering based diarization systems with a relative improvement of 40% and 68%, respectively. We then propose a complete-linkage speaker linking system to achieve attribution and demonstrate a 26% relative improvement in attribution error rate (AER) over the single-linkage speaker linking approach.

Slide attacks on the Sfinks stream cipher

Sfinks is a shift register based stream cipher designed for hardware implementation and submitted to the eSTREAM project. In this paper, we analyse the initialisation process of Sfinks. We demonstrate a slid property of the loaded state of the Sfinks cipher, where multiple key-IV pairs may produce phase shifted keystream sequences. The state update functions of both the initialisation process and keystream generation and also the pattern of the padding affect generation of the slid pairs.

On the probability density function of the product of Rayleigh distributed random variables

In this paper we investigate the distribution of the product of Rayleigh distributed random variables. Considering the Mellin-Barnes inversion formula and using the saddle point approach we obtain an upper bound for the product distribution. The accuracy of this tail-approximation increases as the number of random variables in the product increase.

Characterizing multi-agent team behavior from partial team tracings : evidence from the English Premier League

Real-world AI systems have been recently deployed which can automatically analyze the plan and tactics of tennis players. As the game-state is updated regularly at short intervals (i.e. point-level), a library of successful and unsuccessful plans of a player can be learnt over time. Given the relative strengths and weaknesses of a player’s plans, a set of proven plans or tactics from the library that characterize a player can be identified. For low-scoring, continuous team sports like soccer, such analysis for multi-agent teams does not exist as the game is not segmented into “discretized” plays (i.e. plans), making it difficult to obtain a library that characterizes a team’s behavior. Additionally, as player tracking data is costly and difficult to obtain, we only have partial team tracings in the form of ball actions which makes this problem even more difficult. In this paper, we propose a method to overcome these issues by representing team behavior via play-segments, which are spatio-temporal descriptions of ball movement over fixed windows of time. Using these representations we can characterize team behavior from entropy maps, which give a measure of predictability of team behaviors across the field. We show the efficacy and applicability of our method on the 2010-2011 English Premier League soccer data.

Time-varying bispectral analysis of visually evoked multi-channel EEG

Theoretical foundations of higher order spectral analysis are revisited to examine the use of time-varying bicoherence on non-stationary signals using a classical short-time Fourier approach. A methodology is developed to apply this to evoked EEG responses where a stimulus-locked time reference is available. Short-time windowed ensembles of the response at the same offset from the reference are considered as ergodic cyclostationary processes within a non-stationary random process. Bicoherence can be estimated reliably with known levels at which it is significantly different from zero and can be tracked as a function of offset from the stimulus. When this methodology is applied to multi-channel EEG, it is possible to obtain information about phase synchronization at different regions of the brain as the neural response develops. The methodology is applied to analyze evoked EEG response to flash visual stimulii to the left and right eye separately. The EEG electrode array is segmented based on bicoherence evolution with time using the mean absolute difference as a measure of dissimilarity. Segment maps confirm the importance of the occipital region in visual processing and demonstrate a link between the frontal and occipital regions during the response. Maps are constructed using bicoherence at bifrequencies that include the alpha band frequency of 8Hz as well as 4 and 20Hz. Differences are observed between responses from the left eye and the right eye, and also between subjects. The methodology shows potential as a neurological functional imaging technique that can be further developed for diagnosis and monitoring using scalp EEG which is less invasive and less expensive than magnetic resonance imaging.

Designing a new robust on-line secondary path modeling technique for feedforward active noise control systems

Several approaches have been introduced in the literature for active noise control (ANC) systems. Since the filtered-x least-mean-square (FxLMS) algorithm appears to be the best choice as a controller filter, researchers tend to improve performance of ANC systems by enhancing and modifying this algorithm. This paper proposes a new version of the FxLMS algorithm, as a first novelty. In many ANC applications, an on-line secondary path modeling method using white noise as a training signal is required to ensure convergence of the system. As a second novelty, this paper proposes a new approach for on-line secondary path modeling on the basis of a new variable-step-size (VSS) LMS algorithm in feed forward ANC systems. The proposed algorithm is designed so that the noise injection is stopped at the optimum point when the modeling accuracy is sufficient. In this approach, a sudden change in the secondary path during operation makes the algorithm reactivate injection of the white noise to re-adjust the secondary path estimate. Comparative simulation results shown in this paper indicate the effectiveness of the proposed approach in reducing both narrow-band and broad-band noise. In addition, the proposed ANC system is robust against sudden changes of the secondary path model.

An efficient online secondary path estimation for feedback active noise control systems

In practical cases for active noise control (ANC), the secondary path has usually a time varying behavior. For these cases, an online secondary path modeling method that uses a white noise as a training signal is required to ensure convergence of the system. The modeling accuracy and the convergence rate are increased when a white noise with a larger variance is used. However, the larger variance increases the residual noise, which decreases performance of the system and additionally causes instability problem to feedback structures. A sudden change in the secondary path leads to divergence of the online secondary path modeling filter. To overcome these problems, this paper proposes a new approach for online secondary path modeling in feedback ANC systems. The proposed algorithm uses the advantages of white noise with larger variance to model the secondary path, but the injection is stopped at the optimum point to increase performance of the algorithm and to prevent the instability effect of the white noise. In this approach, instead of continuous injection of the white noise, a sudden change in secondary path during the operation makes the algorithm to reactivate injection of the white noise to correct the secondary path estimation. In addition, the proposed method models the secondary path without the need of using off-line estimation of the secondary path. Considering the above features increases the convergence rate and modeling accuracy, which results in a high system performance. Computer simulation results shown in this paper indicate effectiveness of the proposed method.

Accelerometer-based fetal movement detection

Monitoring fetal wellbeing is a compelling problem in modern obstetrics. Clinicians have become increasingly aware of the link between fetal activity (movement), well-being, and later developmental outcome. We have recently developed an ambulatory accelerometer-based fetal activity monitor (AFAM) to record 24-hour fetal movement. Using this system, we aim at developing signal processing methods to automatically detect and quantitatively characterize fetal movements. The first step in this direction is to test the performance of the accelerometer in detecting fetal movement against real-time ultrasound imaging (taken as the gold standard). This paper reports first results of this performance analysis.

Iris based identity verification robust to sample presentation security attacks

Iris based identity verification is highly reliable but it can also be subject to attacks. Pupil dilation or constriction stimulated by the application of drugs are examples of sample presentation security attacks which can lead to higher false rejection rates. Suspects on a watch list can potentially circumvent the iris based system using such methods. This paper investigates a new approach using multiple parts of the iris (instances) and multiple iris samples in a sequential decision fusion framework that can yield robust performance. Results are presented and compared with the standard full iris based approach for a number of iris degradations. An advantage of the proposed fusion scheme is that the trade-off between detection errors can be controlled by setting parameters such as the number of instances and the number of samples used in the system. The system can then be operated to match security threat levels. It is shown that for optimal values of these parameters, the fused system also has a lower total error rate.

Noise removal from wave-forms and spectrograms derived from natural recordings of the environment

The work described in this technical report is part of an ongoing project to build practical tools for the manipulation, analysis and visualisation of recordings of the natural environment. This report describes the methods we use to remove background noise from spectrograms. It updates techniques previously described in Towsey and Planitz (2011), Technical report: acoustic analysis of the natural environment, downloadable from: http://eprints.qut.edu.au/41131/. It also describes noise removal from wave-forms, a technique not described in the above 2011 technical report.

An algorithm to cluster the spectra in a spectrogram

The work described in this technical report is part of an ongoing project at QUT to build practical tools for the manipulation, analysis and visualisation of recordings of the natural environment. This report describes the algorithm we use to cluster the spectra in a spectrogram. The report begins with a brief description of the signal processing that prepares the spectrograms.

Classifier selection using sequential error ratio criterion for multi-instance and multi-sample fusion

Classifier selection is a problem encountered by multi-biometric systems that aim to improve performance through fusion of decisions. A particular decision fusion architecture that combines multiple instances (n classifiers) and multiple samples (m attempts at each classifier) has been proposed in previous work to achieve controlled trade-off between false alarms and false rejects. Although analysis on text-dependent speaker verification has demonstrated better performance for fusion of decisions with favourable dependence compared to statistically independent decisions, the performance is not always optimal. Given a pool of instances, best performance with this architecture is obtained for certain combination of instances. Heuristic rules and diversity measures have been commonly used for classifier selection but it is shown that optimal performance is achieved for the `best combination performance' rule. As the search complexity for this rule increases exponentially with the addition of classifiers, a measure - the sequential error ratio (SER) - is proposed in this work that is specifically adapted to the characteristics of sequential fusion architecture. The proposed measure can be used to select a classifier that is most likely to produce a correct decision at each stage. Error rates for fusion of text-dependent HMM based speaker models using SER are compared with other classifier selection methodologies. SER is shown to achieve near optimal performance for sequential fusion of multiple instances with or without the use of multiple samples. The methodology applies to multiple speech utterances for telephone or internet based access control and to other systems such as multiple finger print and multiple handwriting sample based identity verification systems.

Histogram of weighted local directions for gait recognition

In this paper, we explore the effectiveness of patch-based gradient feature extraction methods when applied to appearance-based gait recognition. Extending existing popular feature extraction methods such as HOG and LDP, we propose a novel technique which we term the Histogram of Weighted Local Directions (HWLD). These 3 methods are applied to gait recognition using the GEI feature, with classification performed using SRC. Evaluations on the CASIA and OULP datasets show significant improvements using these patch-based methods over existing implementations, with the proposed method achieving the highest recognition rate for the respective datasets. In addition, the HWLD can easily be extended to 3D, which we demonstrate using the GEV feature on the DGD dataset, observing improvements in performance.

Vision-based estimation of airborne target pseudobearing rate using hidden Markov model filters

The problem of estimating pseudobearing rate information of an airborne target based on measurements from a vision sensor is considered. Novel image speed and heading angle estimators are presented that exploit image morphology, hidden Markov model (HMM) filtering, and relative entropy rate (RER) concepts to allow pseudobearing rate information to be determined before (or whilst) the target track is being estimated from vision information.

Speaker attribution of Australian broadcast news data

Speaker attribution is the task of annotating a spoken audio archive based on speaker identities. This can be achieved using speaker diarization and speaker linking. In our previous work, we proposed an efficient attribution system, using complete-linkage clustering, for conducting attribution of large sets of two-speaker telephone data. In this paper, we build on our proposed approach to achieve a robust system, applicable to multiple recording domains. To do this, we first extend the diarization module of our system to accommodate multi-speaker (>2) recordings. We achieve this through using a robust cross-likelihood ratio (CLR) threshold stopping criterion for clustering, as opposed to the original stopping criterion of two speakers used for telephone data. We evaluate this baseline diarization module across a dataset of Australian broadcast news recordings, showing a significant lack of diarization accuracy without previous knowledge of the true number of speakers within a recording. We thus propose applying an additional pass of complete-linkage clustering to the diarization module, demonstrating an absolute improvement of 20% in diarization error rate (DER). We then evaluate our proposed multi-domain attribution system across the broadcast news data, demonstrating achievable attribution error rates (AER) as low as 17%.