Extending the task of diarization to speaker attribution

In this paper we extend the concept of speaker annotation within a single-recording, or speaker diarization, to a collection wide approach we call speaker attribution. Accordingly, speaker attribution is the task of clustering expectantly homogenous intersession clusters obtained using diarization according to common cross-recording identities. The result of attribution is a collection of spoken audio across multiple recordings attributed to speaker identities. In this paper, an attribution system is proposed using mean-only MAP adaptation of a combined-gender UBM to model clusters from a perfect diarization system, as well as a JFA-based system with session variability compensation. The normalized cross-likelihood ratio is calculated for each pair of clusters to construct an attribution matrix and the complete linkage algorithm is employed to conduct clustering of the inter-session clusters. A matched cluster purity and coverage of 87.1% was obtained on the NIST 2008 SRE corpus.

Weighted LDA techniques for I-vector based speaker verification

This paper introduces the Weighted Linear Discriminant Analysis (WLDA) technique, based upon the weighted pairwise Fisher criterion, for the purposes of improving i-vector speaker verification in the presence of high intersession variability. By taking advantage of the speaker discriminative information that is available in the distances between pairs of speakers clustered in the development i-vector space, the WLDA technique is shown to provide an improvement in speaker verification performance over traditional Linear Discriminant Analysis (LDA) approaches. A similar approach is also taken to extend the recently developed Source Normalised LDA (SNLDA) into Weighted SNLDA (WSNLDA) which, similarly, shows an improvement in speaker verification performance in both matched and mismatched enrolment/verification conditions. Based upon the results presented within this paper using the NIST 2008 Speaker Recognition Evaluation dataset, we believe that both WLDA and WSNLDA are viable as replacement techniques to improve the performance of LDA and SNLDA-based i-vector speaker verification.

I-vector based speaker recognition using advanced channel compensation techniques

This paper investigates advanced channel compensation techniques for the purpose of improving i-vector speaker verification performance in the presence of high intersession variability using the NIST 2008 and 2010 SRE corpora. The performance of four channel compensation techniques: (a) weighted maximum margin criterion (WMMC), (b) source-normalized WMMC (SN-WMMC), (c) weighted linear discriminant analysis (WLDA), and; (d) source-normalized WLDA (SN-WLDA) have been investigated. We show that, by extracting the discriminatory information between pairs of speakers as well as capturing the source variation information in the development i-vector space, the SN-WLDA based cosine similarity scoring (CSS) i-vector system is shown to provide over 20% improvement in EER for NIST 2008 interview and microphone verification and over 10% improvement in EER for NIST 2008 telephone verification, when compared to SN-LDA based CSS i-vector system. Further, score-level fusion techniques are analyzed to combine the best channel compensation approaches, to provide over 8% improvement in DCF over the best single approach, (SN-WLDA), for NIST 2008 interview/ telephone enrolment-verification condition. Finally, we demonstrate that the improvements found in the context of CSS also generalize to state-of-the-art GPLDA with up to 14% relative improvement in EER for NIST SRE 2010 interview and microphone verification and over 7% relative improvement in EER for NIST SRE 2010 telephone verification.

Improving short utterance based I-vector speaker recognition using source and utterance-duration normalization techniques

A significant amount of speech is typically required for speaker verification system development and evaluation, especially in the presence of large intersession variability. This paper introduces a source and utterance duration normalized linear discriminant analysis (SUN-LDA) approaches to compensate session variability in short-utterance i-vector speaker verification systems. Two variations of SUN-LDA are proposed where normalization techniques are used to capture source variation from both short and full-length development i-vectors, one based upon pooling (SUN-LDA-pooled) and the other on concatenation (SUN-LDA-concat) across the duration and source-dependent session variation. Both the SUN-LDA-pooled and SUN-LDA-concat techniques are shown to provide improvement over traditional LDA on NIST 08 truncated 10sec-10sec evaluation conditions, with the highest improvement obtained with the SUN-LDA-concat technique achieving a relative improvement of 8% in EER for mis-matched conditions and over 3% for matched conditions over traditional LDA approaches.

Local inter-session variability modelling for object classification

Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based intersession variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset.

Making Ireland Modern

Making Ireland Modern is a cross-disciplinary, inter-institutional, inter-media design and research project which emerged from an open competition (won by Boyd and McLaughlin) to commission/curate the Irish pavilion for the Venice biennale 2014. It explores the relationship between architecture, infrastructure and technology in the building of a new nation. Constructed as a demountable, open matrix of drawings, photographs, models and other artefacts, the exhibition (12 x 5 x 6 metres) presents ten infrastructural episodes – Negation, Electricity, Health, Transportation, Television, Aviation, Education, Telecommunications, Motorways, Data – spanning a period of one hundred years from 1916-2016. Exploring a range of scales from the detail design of objects to entire landscapes and other territories, Making Ireland Modern describes architecture’s role in transforming the physical and cultural identity of the new state through its intersession in the everyday lives of its population. In 2015, we were commissioned by the Arts Council of Ireland to expand and develop the pavilion for a three cities tour of Ireland as one of the five major strands of the Arts Council’s Art2016 programme of Irish State’s 1916-2016 centennial celebrations (2016).

Effects of MRI scan acceleration on brain volume measurement consistency.

PURPOSE: To evaluate the effects of recent advances in magnetic resonance imaging (MRI) radiofrequency (RF) coil and parallel imaging technology on brain volume measurement consistency. MATERIALS AND METHODS: In all, 103 whole-brain MRI volumes were acquired at a clinical 3T MRI, equipped with a 12- and 32-channel head coil, using the T1-weighted protocol as employed in the Alzheimer's Disease Neuroimaging Initiative study with parallel imaging accelerations ranging from 1 to 5. An experienced reader performed qualitative ratings of the images. For quantitative analysis, differences in composite width (CW, a measure of image similarity) and boundary shift integral (BSI, a measure of whole-brain atrophy) were calculated. RESULTS: Intra- and intersession comparisons of CW and BSI measures from scans with equal acceleration demonstrated excellent scan-rescan accuracy, even at the highest acceleration applied. Pairs-of-scans acquired with different accelerations exhibited poor scan-rescan consistency only when differences in the acceleration factor were maximized. A change in the coil hardware between compared scans was found to bias the BSI measure. CONCLUSION: The most important findings are that the accelerated acquisitions appear to be compatible with the assessment of high-quality quantitative information and that for highest scan-rescan accuracy in serial scans the acquisition protocol should be kept as consistent as possible over time. J. Magn. Reson. Imaging 2012;36:1234-1240. ©2012 Wiley Periodicals, Inc.

The interval between the emergence of pharmacologically synchronized ovarian follicular waves and ovum pickup does not significantly affect in vitro embryo production in Bos indicus, Bos taurus, and Bubalus bubalis

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Paradoxical effects of bovine somatotropin treatment on the ovarian follicular population and in vitro embryo production of lactating buffalo donors submitted to ovum pick-up

The aim of the present study was to evaluate the effect of bovine somatotropin (bST; 500 mg) administration on lactating buffalo donors submitted to two different ovum pick-up (OPU) and in vitro embryo production schemes with a 7 or 14 d intersession OPU interval. A total of 16 lactating buffalo cows were randomly assigned into one of four experimental groups according to the bST treatment (bST or No-bST) and the OPU intersession interval (7 or 14 d) in a 2 x 2 factorial design (16 weeks of OPU sessions). The females submitted to OPU every 14d had a larger (P < 0.001) number of ovarian follicles suitable for puncture (15.6 +/- 0.7 vs. 12.8 +/- 0.4) and an increased (P = 0.004) number of cumulus-oocyte complexes (COCs) recovered (10.0 +/- 0.5 vs. 8.5 +/- 0.3) compared to the 7 d interval group. However, a 7 or 14 d interval between OPU sessions had no effect (P = 0.34) on the number of blastocysts produced per OPU (1.0 +/- 0.1 vs. 13 +/- 0.2, respectively). In addition, bST treatment increased (P < 0.001) the number of ovarian follicles suitable for puncture (15.3 +/- 0.5 vs. 12.1 +/- 0.4) but reduced the percentage (18.9% vs. 10.9%; P = 0.009) and the number (1.4 +/- 0.2 vs. 0.8 +/- 0.1; P = 0.003) of blastocysts produced per OPU session compared with the non-bST-treated buffaloes. In conclusion, the 14d interval between OPU sessions and bST treatment efficiently increased the number of ovarian follicles suitable for puncture. However, the OPU session interval had no effect on embryo production, and bST treatment reduced the in vitro blastocyst outcomes in lactating buffalo donors.

Macular thickness measurements in healthy eyes using six different optical coherence tomography instruments

To compare central retinal thickness (CRT) measurements in healthy eyes by different commercially available OCT instruments and to compare the intersession reproducibility of such measurements.

Reproducibility of cerebral phenylalanine levels in patients with phenylketonuria determined by 1H-MR spectroscopy

The reproducibility of metabolite content determined by MR spectroscopy (MRS) is usually at best a few percent for the prominent singlets. When studying low-concentration metabolites, like phenylalanine (Phe), where tissue content can be <100 micromol/kg, better reproducibility is paramount-particularly in view of using MRS results for potential individual treatment advice. An optimized, targeted spectroscopy method was established at 1.5T and reproducibility was established in 21 patients with phenylketonuria (PKU) where three spectra were recorded in each of three independent sessions, two of which were in immediate succession to minimize physiologic variation. Intersession variation was found to be only 7 micromol/kg Phe for back-to-back repetition of sessions, in close agreement with the variation of 16 micromol/kg observed for single spectra within a session. Analysis of variance proved the individuality of the blood/brain Phe ratio-though this ratio seems to be influenced by physiologic factors that are not stable in time. The excellent reproducibility was achieved through optimization of various factors, including signal-to-noise ratio, repositioning, and prescan calibrations, but also by enforcing as much prior information as possible (e.g., lineshape and phase from reference scans, constant prior-knowledge-locked baseline). While the application of maximum general prior knowledge is a general method to reduce fluctuations, one should remember that it may introduce systematic errors.

Clinical evaluation of the Shin-Nippon NVision-K 5001/Grand Seiko WR-5100K autorefractor

Purpose. A clinical evaluation of the Shin-Nippon NVision-K 5001 (also branded as the Grand Seiko WR-5100K) autorefractor (Japan) was performed to examine validity and repeatability compared with subjective refraction and Javal-Schiotz keratometry. Methods. Measurements of refractive error were performed on 198 eyes of 99 subjects (aged 23.2 ± 7.4 years) subjectively (noncycloplegic) by one masked optometrist and objectively with the NVision-K autorefractor by a second optometrist. Keratometry measurements using the NVision-K were compared with the Javal-Schiotz keratometer. Intrasession repeatability of the NVision-K was also assessed on all 99 subjects together with intersession repeatability on a separate occasion separated by 7 to 14 days. Results. Refractive error as measured by the NVision-K was found to be similar (p = 0.67) to subjective refraction (difference, 0.14 ± 0.35 D). It was both accurate and repeatable over a wide prescription range (-8.25 to +7.25 D). Keratometry as measured by the NVision-K was found to be similar (p > 0.50) to the Javal-Schiotz technique in both the horizontal and vertical meridians (horizontal: difference, 0.02 ± 0.09 mm; vertical: difference, 0.01 ± 0.14 mm). There was minimal bias, and the results were repeatable (horizontal: intersession difference, 0.00 ± 0.09 mm; vertical: intersession difference, -0.01 ± 0.12 mm). Conclusion. The open-view arrangement of the Shin-Nippon NVision-K 5001 facilitates the measurement of static refractive error and the accommodative response to real-world stimuli. Coupled with its accuracy, repeatability, and capability to measure corneal curvature, it is a valuable addition to objective instrumentation currently available to the optometrist and researcher.

Operator-induced errors in Hartmann-Shack wavefront sensing:model eye study

Purpose: To evaluate the effects of instrument realignment and angular misalignment during the clinical determination of wavefront aberrations by simulation in model eyes. Setting: Aston Academy of Life Sciences, Aston University, Birmingham, United Kingdom. Methods: Six model eyes were examined with wavefront-aberration-supported cornea ablation (WASCA) (Carl Zeiss Meditec) in 4 sessions of 10 measurements each: sessions 1 and 2, consecutive repeated measures without realignment; session 3, realignment of the instrument between readings; session 4, measurements without realignment but with the model eye shifted 6 degrees angularly. Intersession repeatability and the effects of realignment and misalignment were obtained by comparing the measurements in the various sessions for coma, spherical aberration, and higher-order aberrations (HOAs). Results: The mean differences between the 2 sessions without realignment of the instrument were 0.020 μm ± 0.076 (SD) for Z3 - 1(P = .551), 0.009 ± 0.139 μm for Z3 1(P = .877), 0.004 ± 0.037 μm for Z4 0 (P = .820), and 0.005 ± 0.01 μm for HO root mean square (RMS) (P = .301). Differences between the nonrealigned and realigned instruments were -0.017 ± 0.026 μm for Z3 - 1(P = .159), 0.009 ± 0.028 μm for Z3 1 (P = .475), 0.007 ± 0.014 μm for Z4 0(P = .296), and 0.002 ± 0.007 μm for HO RMS (P = 0.529; differences between centered and misaligned instruments were -0.355 ± 0.149 μm for Z3 - 1 (P = .002), 0.007 ± 0.034 μm for Z3 1(P = .620), -0.005 ± 0.081 μm for Z4 0(P = .885), and 0.012 ± 0.020 μm for HO RMS (P = .195). Realignment increased the standard deviation by a factor of 3 compared with the first session without realignment. Conclusions: Repeatability of the WASCA was excellent in all situations tested. Realignment substantially increased the variance of the measurements. Angular misalignment can result in significant errors, particularly in the determination of coma. These findings are important when assessing highly aberrated eyes during follow-up or before surgery. © 2007 ASCRS and ESCRS.