Speaker diarization: Segmentation and clustering of speeches

Speaker diarization is the process of sorting speeches according to the speaker. Diarization helps to search and retrieve what a certain speaker uttered in a meeting. Applications of diarization systemsextend to other domains than meetings, for example, lectures, telephone, television, and radio. Besides, diarization enhances the performance of several speech technologies such as speaker recognition, automatic transcription, and speaker tracking. Methodologies previously used in developing diarization systems are discussed. Prior results and techniques are studied and compared. Methods such as Hidden Markov Models and Gaussian Mixture Models that are used in speaker recognition and other speech technologies are also used in speaker diarization. The objective of this thesis is to develop a speaker diarization system in meeting domain. Experimental part of this work indicates that zero-crossing rate can be used effectively in breaking down the audio stream into segments, and adaptive Gaussian Models fit adequately short audio segments. Results show that 35 Gaussian Models and one second as average length of each segment are optimum values to build a diarization system for the tested data. Uniting the segments which are uttered by same speaker is done in a bottom-up clustering by a newapproach of categorizing the mixture weights.

Assessing Organizational Cultural Fit prior to Integration in the M&A Process

Yritysostojen määrä on historiallisen suuri 2000-luvulla, vaikka melkein puolet niistä epäonnistuu. Aineettomilla tekijöillä, kuten organisaatiokulttuureilla, on keskeinen rooli yritysostojen onnistumisissa. Myös case yritys on aktiivinen yritysostoissa ja haluaa arvioida integraatioprosessinsa tehokkuutta. Siten diplomityön tarkoituksena on luoda työkalu organisaatiokulttuurien yhteensopivuuden arvioimiseksi, jotta ostopäätöksentekoa sekä integraation suunnittelua voitaisiin tukea paremmin yrityksessä. Diplomityö vastaakin kysymyksiin, kuten miten arvioida kulttuurista yhteensopivuutta ennen integraatiota integraatioprosessin parantamiseksi sekä mitkä ovat olleet kaikkein ongelmallisimmat ja toisaalta kaikkein menestyksekkäimmät kulttuuritekijät tutkitussa integraatiossa. Kulttuurisen yhteensopivuuden arviointi tulisi nähdä prosessina osana yrityskauppaa. Prosessin tulisi alkaa kulttuurisen integraation tavoitteiden määrittämisellä sekä organisaatiokulttuurin käsitteen ymmärtämisellä. Kulttuurianalyysi tulisi suorittaa työpajan avulla. Sen tulisi käsitellä ainakin yhdeksän kulttuurin osa-aluetta: innovatiivisuus, päätöksenteko, ihmissuuntautuneisuus, kommunikaatio, kontrolli, asiakassuuntautuneisuus, ajanhallinta, identifikaatio, sekä kollektivismi. Lisäksi kuhunkin dimensioon liittyvään kysymykseen tulisi vastata pisteillä yhdestä viiteen, jolloin voidaan piirtää kulttuurisen yhteensopivuuden kuvio. Tämän jälkeen johdon tulisi keskustella tuloksista vielä kerran tarkemmin ja lopulta koota tulokset kirjalliseksi raportiksi. Tutkitussa integraatiossa parhaiten integraatiota tukivat ihmissuuntautuneisuus sekä ajanhallinta (työn ja vapaa-ajan välinen tasapaino sekä tulevaisuus-suuntautuneisuus). Haasteellisimmat kulttuuritekijät koskivat päätöksentekoa, kommunikaatiota ja kontrollia, jotka vaikuttavat olevan tyypillisiä ongelmia ison yrityksen ostaessa pienemmän yrityksen.

Winning time in land warfare : a reading of the classical strategists prior to the age of mechanization

We live in an age where rationalization and demands of efficiency taint every aspect of our lives both as individuals and as a society. Even warfare cannot escape the increased speed of human interaction. Time is a resource to be managed. It has to be optimized, saved and won in military affairs as well. The purpose of this research paper is to analyze the dogmatic texts of military thought to search for answers what the classics of strategy saw in the interrelations of temporality and warfare and if their thoughts remain meaningful in the contemporary conjunction. Since the way a society functions is reflected in the way it conducts its wars, there naturally are differences between an agrarian, industrial and information society. Theorists of different eras emphasize things specific to their times, but warfare, like any human interaction, is always bounded by temporality. Not only is the pace of warfare dependent on the progress of the society, but time permeates warfare in all its aspects. This research paper focuses on two specific topics that arose from the texts themselves; how should time be managed and manipulated in warfare and how to economize and “win” it from the enemy. A method where lengthy quotations are used to illustrate the main point of the strategists has been chosen for this research paper. While Clausewitz is the most prominent source of quotations, thoughts from ancient India and China are represented as well to prove that the combination of right force in the right place at the right time is still the way of the victorious. Tactics change in the course of time but the principles of strategy remain unaltered and are only adapted to suit new situations. While ancient and pre-modern societies had their focus on finding auspicious moments for battle in the flow of kronos-time based on divinities, portents and auguries, we can trace elements of manipulation of time in warfare from the earliest surviving texts. While time as a fourth dimension of the battlespace emerged only in the modern era, all through the history of military thought it has had a profound meaning. In the past time could be squandered, today it always has to be won. This paper asks the question “why”.

Weed Community in a Raw Sugarcane Renovation Area Submitted to Different Soil Managements

The objective of this study was to evaluate the weed community in a raw sugarcane renovation area with three soil managements and peanut sowing in succession. The experiment was conducted during the 2007/08 season on a raw sugarcane area harvested without prior burning in the last five cuts. A randomized block design with treatments arranged in a split plot and arranged in four replications was used. The main treatments consisted of three cropping systems: conventional tillage, minimum tillage and direct planting, and subplots consisted in the absence (resting) or presence of crop rotation with peanuts. After 135 days from planting peanuts and 180 days of sugarcane harvest, the number of weeds m-2 was counted and the shoot dry biomass of the weeds collected was determined. The data were interpreted by analysis of variance and the means were compared by Tukey's test at 5% probability so that phytosociological indices a, b, c e d were calculated. The use of soil conservation tillage and peanut in rotation with sugarcane in the renovation areas is effective in controlling weeds and suppression of weed species difficult to control like Cyperus rotundus, Commelina bengalensis, Urochloa plantaginea, and Digitaria nuda.

Context-specific independence in graphical models

The theme of this thesis is context-speci c independence in graphical models. Considering a system of stochastic variables it is often the case that the variables are dependent of each other. This can, for instance, be seen by measuring the covariance between a pair of variables. Using graphical models, it is possible to visualize the dependence structure found in a set of stochastic variables. Using ordinary graphical models, such as Markov networks, Bayesian networks, and Gaussian graphical models, the type of dependencies that can be modeled is limited to marginal and conditional (in)dependencies. The models introduced in this thesis enable the graphical representation of context-speci c independencies, i.e. conditional independencies that hold only in a subset of the outcome space of the conditioning variables. In the articles included in this thesis, we introduce several types of graphical models that can represent context-speci c independencies. Models for both discrete variables and continuous variables are considered. A wide range of properties are examined for the introduced models, including identi ability, robustness, scoring, and optimization. In one article, a predictive classi er which utilizes context-speci c independence models is introduced. This classi er clearly demonstrates the potential bene ts of the introduced models. The purpose of the material included in the thesis prior to the articles is to provide the basic theory needed to understand the articles.

Effects of metformin treatment on luteal phase progesterone concentration in polycystic ovary syndrome

The causes of luteal phase progesterone deficiency in polycystic ovary syndrome (PCOS) are not known. To determine the possible involvement of hyperinsulinemia in luteal phase progesterone deficiency in women with PCOS, we examined the relationship between progesterone, luteinizing hormone (LH) and insulin during the luteal phase and studied the effect of metformin on luteal progesterone levels in PCOS. Patients with PCOS (19 women aged 18-35 years) were treated with metformin (500 mg three times daily) for 4 weeks prior to the test cycle and throughout the study period, and submitted to ovulation induction with clomiphene citrate. Blood samples were collected from control (N = 5, same age range as PCOS women) and PCOS women during the late follicular (one sample) and luteal (3 samples) phases and LH, insulin and progesterone concentrations were determined. Results were analyzed by one-way analysis of variance (ANOVA), Duncan's test and Karl Pearson's coefficient of correlation (r). The endocrine study showed low progesterone level (4.9 ng/ml) during luteal phase in the PCOS women as compared with control (21.6 ng/ml). A significant negative correlation was observed between insulin and progesterone (r = -0.60; P < 0.01) and between progesterone and LH (r = -0.56; P < 0.05) concentrations, and a positive correlation (r = 0.83; P < 0.001) was observed between LH and insulin. The study further demonstrated a significant enhancement in luteal progesterone concentration (16.97 ng/ml) in PCOS women treated with metformin. The results suggest that hyperinsulinemia/insulin resistance may be responsible for low progesterone levels during the luteal phase in PCOS. The luteal progesterone level may be enhanced in PCOS by decreasing insulin secretion with metformin.

Relationship of inflammatory markers and pain in patients with head and neck cancer prior to anticancer therapy

Pain is a common symptom in patients with cancer, including those with head and neck cancer (HNC). While studies suggest an association between chronic inflammation and pain, levels of inflammatory cytokines, such as C-reactive protein (CRP) and tumor necrosis factor-alpha (TNF-α), have not been correlated with pain in HNC patients who are not currently undergoing anticancer treatment. The purpose of this study was to examine the relationship between these inflammatory markers and perceived pain in HNC patients prior to anticancer therapy. The study group consisted of 127 HNC patients and 9 healthy controls. Pain was assessed using the Brief Pain Inventory (BPI), and serum levels of CRP and TNF-α were determined using the particle-enhanced turbidimetric immunoassay (PETIA) and ELISA techniques, respectively. Patients experiencing pain had significantly higher levels of CRP (P<0.01) and TNF-α (P<0.05) compared with controls and with patients reporting no pain. There were significantly positive associations between pain, CRP level, and tumor stage. This is the first study to report a positive association between perceived pain and CRP in HNC patients at the time of diagnosis. The current findings suggest important associations between pain and inflammatory processes in HNC patients, with potential implications for future treatment strategies.

Voluntary exercise prior to traumatic brain injury alters miRNA expression in the injured mouse cerebral cortex

MicroRNAs (miRNAs) may be important mediators of the profound molecular and cellular changes that occur after traumatic brain injury (TBI). However, the changes and possible roles of miRNAs induced by voluntary exercise prior to TBI are still not known. In this report, the microarray method was used to demonstrate alterations in miRNA expression levels in the cerebral cortex of TBI mice that were pretrained on a running wheel (RW). Voluntary RW exercise prior to TBI: i) significantly decreased the mortality rate and improved the recovery of the righting reflex in TBI mice, and ii) differentially changed the levels of several miRNAs, upregulating some and downregulating others. Furthermore, we revealed global upregulation of miR-21, miR-92a, and miR-874 and downregulation of miR-138, let-7c, and miR-124 expression among the sham-non-runner, TBI-non-runner, and TBI-runner groups. Quantitative reverse transcription polymerase chain reaction data (RT-qPCR) indicated good consistency with the microarray results. Our microarray-based analysis of miRNA expression in mice cerebral cortex after TBI revealed that some miRNAs such as miR-21, miR-92a, miR-874, miR-138, let-7c, and miR-124 could be involved in the prevention and protection afforded by voluntary exercise in a TBI model.

Is a contrast study really necessary prior to ureteroscopy?

This study aimed to evaluate the effect of preoperative imaging techniques on the success and complication rates of ureteroscopy. We performed a retrospective analysis of 736 patients (455 males and 281 females), with a mean age of 45.5±15.2 years (range, 1-88 years), who underwent rigid ureteroscopic procedures for removal of ureteral stones. Patients were divided into 4 groups according to the type of imaging modality used: group I, intravenous urography (n=116); group II, computed tomography (n=381); group III, computed tomography and intravenous urography (n=91), and group IV, ultrasonography and abdominal plain film (n=148). Patients’ demographics, stone size and location, prior shock wave lithotripsy, lithotripsy technique, operation time, success rate, and rate of intraoperative complications were compared among the groups. There were no significant differences in success and complication rates among the groups. The stone-free rate after primary ureteroscopy was 87.1% in group I, 88.2% in group II, 96.7% in group III, and 89.9% in group IV (P=0.093). The overall incidence of intraoperative complications was 11.8%. According to the modified Satava classification system, 6.1% of patients had grade 1, 5.1% had grade 2, and 0.54% had grade 3 complications. Intraoperative complications developed in 12.1% of patients in group I, 12.6% of patients in group II, 7.7% of patients in group III, and 12.2% of patients in group IV (P=0.625). Our findings clearly demonstrate that ureteroscopic treatment of ureteral stones can be safely and effectively performed with no use of contrast study imaging, except in doubtful cases of anatomical abnormalities.

High-impact exercise in rats prior to and during suspension can prevent bone loss

High-impact exercise has been considered an important method for treating bone loss in osteopenic experimental models. In this study, we investigated the effects of osteopenia caused by inactivity in femora and tibiae of rats subjected to jump training using the rat tail suspension model. Eight-week-old female Wistar rats were divided into five groups (n=10 each group): jump training for 2 weeks before suspension and training during 3 weeks of suspension; jump training for 2 weeks before suspension; jump training only during suspension; suspension without any training; and a control group. The exercise protocol consisted of 20 jumps/day, 5 days/week, with a jump height of 40 cm. The bone mineral density of the femora and tibiae was measured by double energy X-ray absorptiometry and the same bones were evaluated by mechanical tests. Bone microarchitecture was evaluated by scanning electron microscopy. One-way ANOVA was used to compare groups. Significance was determined as P<0.05. Regarding bone mineral density, mechanical properties and bone microarchitecture, the beneficial effects were greater in the bones of animals subjected to pre-suspension training and subsequently to training during suspension, compared with the bones of animals subjected to pre-suspension training or to training during suspension. Our results indicate that a period of high impact exercise prior to tail suspension in rats can prevent the installation of osteopenia if there is also training during the tail suspension.

Optimizing counterparty credit risk portfolio by reducing exposure at default variance

Rahoitusyhtiöt pitävät omaa pääomaa taseessa harvinaisen suuria markkinamuutoksia varten ja tämän pääoman määrä on ohjattu valvontaviranomaisten toimesta. Euroopassa Basel akordi antaa suuntaviivat näille säädöksille. 2007 vuoden finanssikriisin jälkeen rahoitusyhtiöt sekä valvojat ovat olleet entistä kiinnostuneempia pääoman riittävyydestä. Tutkimuksia riskistä, säädöksistä ja pääomavaateen madaltamisesta on tehty aikaisemmin, mutta tässä tutkimuksessa keskitytään vaateen jatkuvan muutoksen suuruuteen. Tutkimus sisältää hypoteettisen vastapuoliriski portfolion, jossa on mukana valuuttajohdannaisia. Tätä portfoliota simuloidaan eri tavoin, jotta nähtäisiin kuinka suuri vaikutus portfolion koostumuksella voi olla pääomavaateen varianssiin. Jos tämä muuttuja on merkittävä, pitäisikö rahoitusyhtiöiden yrittää pienentää muutosta, jotta yhtiöiden varapääoman määrää voitaisiin alentaa? Tutkimuksessa on myös haastateltu Suomen johtavia vastapuoliriski asiantuntijoita, jotta nähtäisiin rahoitusalan oma näkemys asian merkittävyydestä. Tutkimusmenetelminä toimivat haastattelut sekä numeerinen analyysi hypoteettisella portfoliolla. Kaupat tähän vastapuoliriski portfolioon on luotu 14 vuoden ajalle ja se sisältää ainoastaan valuuttajohdannaisia viidessä eri valuutassa. Riski lasketaan markkina-arvo menetelmällä, joista lasketaan VaR-mallilla tulevaisuuden riski nettoutuksen kera. Portfolion rakennetta muutetaan simuloinneissa, jotta nähtäisiin vaikutus tulevaisuuden riskeille, joita käytetään edustamaan pääomavaateen määrää ja sen vaihtelua yli ajan. Portfolioiden riskejä lasketaan myös rasituskokeiden avulla, jotta tuloksista saataisiin mahdollisimman todenmukaisia. Analyyttinen osuus tutkimuksesta näyttää sen, että tämän kaltainen optimointi on suuresti riippuvainen alkuperäisestä portfoliosta, jonka määrittää yleisesti rahoitusyhtiön myyntistrategia. Yleisesti ottaen pääomavaateen varianssin muutos voi simuloinneissa olla melko suurta, varsinkin jos mukaan huomioidaan rasitus testit, puuttuvat tuotteet sekä muut pääomavaateen laskentaan huomioitavat seikat. Haastatteluissa saatiin selville millainen optimointi voisi olla mahdollista todellisuudessa. Huomattiin myös että tämän kaltainen ajattelumalli on jo huomattu alalla ennestään. Jon Gregory jopa mainitsi, että jotkin rahoitusyhtiöt ovat enemmän kiinnostuneita muutosten pienentämisestä kuin itse pääomavaateen suuruudesta. Näyttääkin siltä, että tämän aihepiiri vaatisi entistä enemmän tutkimusta, sillä sitä ei ennestään vielä ole, ja rahoitusyhtiöt ovat jo alkaneet etsimään uusia keinoja selvitäkseen rahoitusalalla, joka on yhä entisestään kilpailullisempi.

Effect of holding period prior to storage on the chemical attributes of Starking Delicious apples during refrigerated storage

In this research, the effects of three different holding periods (6, 12 and 24 hours) prior to storage on the quality attributes of Starking Delicious apples were investigated during storage of 8 months at 0.5 ± 1.0 ºC. Changes in weight loss, flesh firmness, pH values, soluble dry matter amount, titratable acidity values, ascorbic acid contents, and total and reducing sugar content were determined. According to the results, the holding period showed statistically significant changes in the quality attributes of the apples (p < 0.05).

Spectral Analysis of Symmetric and Anti-Symmetric Pairwise Kernels

This work investigates theoretical properties of symmetric and anti-symmetric kernels. First chapters give an overview of the theory of kernels used in supervised machine learning. Central focus is on the regularized least squares algorithm, which is motivated as a problem of function reconstruction through an abstract inverse problem. Brief review of reproducing kernel Hilbert spaces shows how kernels define an implicit hypothesis space with multiple equivalent characterizations and how this space may be modified by incorporating prior knowledge. Mathematical results of the abstract inverse problem, in particular spectral properties, pseudoinverse and regularization are recollected and then specialized to kernels. Symmetric and anti-symmetric kernels are applied in relation learning problems which incorporate prior knowledge that the relation is symmetric or anti-symmetric, respectively. Theoretical properties of these kernels are proved in a draft this thesis is based on and comprehensively referenced here. These proofs show that these kernels can be guaranteed to learn only symmetric or anti-symmetric relations, and they can learn any relations relative to the original kernel modified to learn only symmetric or anti-symmetric parts. Further results prove spectral properties of these kernels, central result being a simple inequality for the the trace of the estimator, also called the effective dimension. This quantity is used in learning bounds to guarantee smaller variance.