Wyniki wyszukiwania - Biblioteka Nauki

1

Wybrane metody pomiaru magnetostrykcji materiałów wykazujących gigantyczną magnetostrykcję (GMM, ang. Giant Magnetostrictive Materials)

100%

Tomiczek A. , Żuberek R. , Mech R.

LAB Laboratoria, Aparatura, Badania

|

2019

|

tom R. 24, nr 6

18--21

2

Testing for the economic and environmental impacts of EU Emissions Trading System: A panel GMM approach

100%

Gretszel P. , Gurgul H. , Lach Ł. , Schleicher S.

Managerial Economics

|

2020

|

tom Vol. 21, no. 2

99--125

EN

The COVID-19 pandemic has had a great impact on the economies of the EU, also with regard to the future of EU climate policy. The plan to rebuild and support the EU economy seems to place less emphasis on environmental issues as the main focus has been shifted to a quick economic recovery. One of the issues discussed in this context is the continued operation of the EU ETS. From this perspective, empirical research devoted to a thorough analysis of the impact of the EU ETS is of particular importance. At the same time, the current economic literature lacks any econometric analyzes devoted to the issues in question that would use detailed and reliable databases on EU ETS like the one provided by the Wegener Center for Climate and Global Change. The aim of this paper is to make a preliminary assessment of the effectiveness of the EU ETS in terms of reducing the actual emissions while preserving the economic growth of EU member states. The extensive empirical analysis is focused on examining the issues in question for different phases of the EU ETS and various groups of EU economies that vary in terms of economic development and the overall air pollutant emission.

3

Testing for the economic and environmental impacts of EU Emissions Trading System: A panel GMM approach

94%

Gretszel P. , Gurgul H. , Lach Ł. , Sleicher S.

Managerial Economics

|

2020

|

tom 21

|

nr 2

99-125

XX

The COVID‑19 pandemic has had a great impact on the economies of the EU, also with regard to the future of EU climate policy. The plan to rebuild and support the EU economy seems to place less emphasis on environmental issues as the main focus has been shifted to a quick economic recovery. One of the issues discussed in this context is the continued operation of the EU ETS. From this perspective, empirical research devoted to a thorough analysis of the impact of the EU ETS is of particular importance. At the same time, the current economic literature lacks any econometric analyzes devoted to the issues in question that would use detailed and reliable databases on EU ETS like the one provided by the Wegener Center for Climate and Global Change. The aim of this paper is to make a preliminary assessment of the effectiveness of the EU ETS in terms of reducing the actual emissions while preserving the economic growth of EU member states. The extensive empirical analysis is focused on examining the issues in question for different phases of the EU ETS and various groups of EU economies that vary in terms of economic development and the overall air pollutant emission.

4

Real time recognition of speakers from internet audio stream

94%

Weychan R. , Marciniak T. , Stankiewicz A. , Dabrowski A.

Foundations of Computing and Decision Sciences

|

2015

|

tom Vol. 40, No. 3

223--233

EN

In this paper we present an automatic speaker recognition technique with the use of the Internet radio lossy (encoded) speech signal streams. We show an influence of the audio encoder (e.g., bitrate) on the speaker model quality. The model of each speaker was calculated with the use of the Gaussian mixture model (GMM) approach. Both the speaker recognition and the further analysis were realized with the use of short utterances to facilitate real time processing. The neighborhoods of the speaker models were analyzed with the use of the ISOMAP algorithm. The experiments were based on four 1-hour public debates with 7–8 speakers (including the moderator), acquired from the Polish radio Internet services. The presented software was developed with the MATLAB environment.

5

DOES CAPITAL FLIGHT HARM DOMESTIC INVESTMENT? AN EMPIRICAL INVESTIGATION FROM A PANEL OF EMERGING AND ADVANCED EUROPE

94%

Workie Tiruneh M. , Siranova M. , Ostrihon F.

Ekonomický časopis (Journal of Economics)

|

2021

|

tom 69

|

nr 2

135 – 157

EN

In this study, we estimate the size of capital flight and its spill over effect on domestic investment in selected European economies during the 2006 – 2016 period. The results based on the fixed effects ordinary least squares method suggest that capital flight has an adverse impact on investment in the economies included in our sample. As a robustness check, we also use the system generalized method of moments (GMM). Our results based on a mix of emerging and advanced economies are fairly similar with previous studies based on emerging economies alone (Yalta, 2010) and developing economies (Ndiaye, 2009).

6

An embedded system for real-time speaker recognition using Raspberry Pi platform

84%

Weychan R. , Marciniak T. , Dąbrowski A.

Elektronika : konstrukcje, technologie, zastosowania

|

2016

|

tom Vol. 57, nr 4

3--6

EN

The paper presents an embedded system, which realizes real time speaker recognition from the internet radio broadcasts. The proposed solution was developed with the use of the open source Python programming language. It was first tested within the Windows environment, then adapted to the Unix operating system in order to use is on the Raspberry Pi 2 platform. We analyzed available libraries to select the most convenient solutions for individual blocks of the speaker recognition task. In the paper we also indicate parameters, for which the algorithm exhibits the greatest efficiency. The prepared software is available on the Github file repository.

PL

Artykuł prezentuje system realizujący rozpoznawanie mówcy z radia internetowego. Zaproponowane rozwiązanie wykorzystuje narzędzia udostępnione w ramach ogólnie dostępnego oprogramowania dla języka Python. Prezentowane oprogramowanie zostało przetestowane w środowisku Windows a następnie zostało zaadaptowane do uruchomienia na platformie Raspberry Pi 2, zarządzanej przez system Linux. W artykule przeanalizowano dostępne biblioteki, które posłużyły do implementacji algorytmów ekstrakcji cech oraz modelowania sygnału mowy. Przeprowadzone eksperymenty pozwoliły na dobranie parametrów systemu, przy których uzyskuje się najlepszą skuteczność identyfikacji i jednocześnie największą szybkość przetwarzania danych. Przygotowane oprogramowanie jest dostępne w repozytorium Github.

7

Analyzing the historical and current perspective of the relationship between inventory policies and CSR

84%

Mohammad S. J. , Saxunová D. , George K. B. , Comite U. , Ahmed M. , Popp J.

Polish Journal of Management Studies

|

2023

|

tom Vol. 28, No. 2

223--238

EN

In this study, we analyze how a company's inventory policy changes because of adopting CSR policies in the manufacturing companies listed on the Shanghai Shenzhen stock exchange by targeting manufacturing companies, as their overall policies are greatly impacted by the level of inventory. The analysis focus on how inventory policies affect Chinese companies' CSR. Inventory levels affect company policies therefore we chose manufacturing companies. The uniqueness of the topic warranted us to research CSR and Inventory management from historical and current viewpoints. The topic has few investigations, especially in China. Increased worker, consumer, and environmental satisfaction may impact volume, according to this study. Given the uniqueness of the topic, there are a very limited number of studies conducted on the topic, especially in the Chinese context, which warrants us to conduct the study, keeping in view the historical and present perspectives of both CSR and Inventory management. In this study, we take this step further by examining whether the volume is affected by efforts to improve the satisfaction of staff, consumers, and the natural environment. We found that customer satisfaction positively affects the cost-to-sales ratio. Environmental consciousness harms the ratio, and employee satisfaction has an inverted U-shaped association with this ratio. It was also found that customer satisfaction positively correlates with inventory levels. Our research has shown that increased levels of social responsibility led to decreased levels of customer corporate social responsibility and increased levels of CSR, which explains the non-linear relationship between a company's total CSR and the to-sales ratio. According to the collected data, the ratio of profits to sales tends to increase when a company's CSR rating is much higher (or lower) than the mean of the CSR distribution.

PL

W niniejszym badaniu analizujemy, jak zmienia się polityka firmy w zakresie zapasów w związku z przyjęciem polityki CSR w spółkach produkcyjnych notowanych na giełdzie w Szanghaju i Shenzhen. Skoncentrowano się na spółkach produkcyjnych, ponieważ ich profil działania i ogólna polityka ma duży wpływ na poziom zapasów. Analiza koncentruje się na tym, jak polityka dotycząca zapasów wpływa na CSR chińskich firm. Wyjątkowość tematu skłoniła nas do zbadania CSR i zarządzania zapasami z historycznego i aktualnego punktu widzenia. Temat ten obejmuje niewiele badań, zwłaszcza w Chinach. Zgodnie z niniejszym badaniem zwiększona satysfakcja pracowników, konsumentów i środowiska może mieć wpływ na wielkość produkcji. Biorąc pod uwagę wyjątkowość tematu, istnieje bardzo ograniczona liczba badań przeprowadzonych na ten temat, zwłaszcza w kontekście chińskim, co uzasadnia przeprowadzenie badania, biorąc pod uwagę historyczne i obecne perspektywy zarówno CSR, jak i zarządzania zapasami. W niniejszym badaniu poszliśmy o krok dalej, sprawdzając, czy wysiłki na rzecz poprawy satysfakcji pracowników, konsumentów i środowiska naturalnego mają wpływ na wielkość sprzedaży. Okazało się, że zadowolenie klientów pozytywnie wpływa na stosunek kosztów do sprzedaży. Świadomość ekologiczna szkodzi temu wskaźnikowi, a zadowolenie pracowników ma związek z tym wskaźnikiem w kształcie odwróconej litery U. Stwierdzono również, że zadowolenie klientów pozytywnie koreluje z poziomem zapasów. Nasze badania wykazały, że zwiększony poziom odpowiedzialności społecznej doprowadził do zmniejszenia poziomu społecznej odpowiedzialności biznesu klientów i zwiększenia poziomu CSR, co wyjaśnia nieliniowy związek między całkowitą CSR firmy a wskaźnikiem sprzedaży. Zgodnie z zebranymi danymi, stosunek zysków do sprzedaży ma tendencję wzrostową, gdy ocena CSR firmy jest znacznie wyższa (lub niższa) niż średnia rozkładu CSR.

8

How Does Transparency Affect Bank Risk and Performance? Evidence from Turkey

84%

Şahan M. , Eksi I. H. , Hacievliyagil N.

Problemy Zarządzania

|

2022

|

tom 20

|

nr 3(97)

148-168

PL

Cel: głównym celem opracowania jest zbadanie wpływu przejrzystości na wyniki banków, które należą do najważniejszych jednostek sektora finansowego. Metodologia: do analizy z wykorzystaniem uogólnionej metody momentów (Generalized Method of Moments, GMM) wykorzystano dane roczne z 22 banków depozytowych działających w Turcji. Opracowano cztery modele związane z rentownością, ryzykiem kredytowym, depozytami i stopami zwrotu z akcji. Dokonano tego poprzez obliczenie wyniku oceny przejrzystości uzyskanego na podstawie 106 kryteriów dla każdego roku i dla każdego banku. Wyniki: zgodnie z wynikami GMM zaobserwowano, że przejrzystość, ryzyko kredytowe i rentowność były ze sobą ujemnie skorelowane, podczas gdy stopy zwrotu z akcji wykazały związek pozytywny. Ograniczenia badawcze: liczba banków notowanych na giełdzie, zwłaszcza w sekcji stóp zwrotu z akcji, jest niewystarczająca. Chociaż badana próba była największa spośród prób wykorzystanych w dotychczasowych badaniach, nie było możliwości objęcia jej zakresem wszystkich banków działających w Turcji. Wartość: analiza wykazuje znaczenie raportowania i udostępniania informacji z banków. Banki powinny określić kryterium przejrzystości, a wynik oceny przejrzystości należy ustalić z wykorzystaniem badanego kryterium.

EN

Purpose: The main objective of this study is to investigate the effect of transparency on the performance of banks, which are among the most important units of the financial sector. Methodology: The Generalized Method of Moments (GMM) analysis was applied using the annual data from 22 deposit banks operating in Turkey. Four models related to profitability, credit risk, deposits, and stock returns were established by calculating a transparency score derived on the basis of 106 criteria for each year and for each bank. Findings: According to the GMM results, it was observed that transparency, credit risk, and profitability were negatively correlated, while stock returns had a positive relationship. Research limitations: There are not enough public-traded banks, especially in the stock returns section. Although this research has the largest sample size among the studies conducted to date, all banks in Turkey could not be included in its scope. Value: The analysis reveals the importance of reporting and sharing information from banks. Banks should set a transparency criterion, and a transparency score should be established using the researched criterion.

9

Speaker recognition based on telephone quality short Polish sequences with removed silence

84%

Marciniak. T. , Krzykowska A. , Weychan R.

Przegląd Elektrotechniczny

|

2012

|

tom R. 88, nr 6

42-46

EN

This paper presents the effectiveness of speaker identification based on short Polish sequences. An impact of automatic removal of silence on the speaker recognition accuracy is considered. Several methods to detect the beginnings and ends of the voice signal have been used. Experimental research was carried out in Matlab environment with the use of a specially prepared database of short speech sequences in Polish. The construction of speaker models was realized with two techniques: Vector Quantization (VQ) and Gaussian Mixture Models (GMM). We also tested the influence of the sampling rate reduction on the speaker recognition performance.

PL

Artykuł przedstawia badania efektywności rozpoznawania mówcy opartego na krótkich wypowiedziach w języku polskim. Sprawdzono wpływ automatycznego wykrywania i usuwania ciszy na jakość rozpoznawania mówcy. Przebadano kilka różnych metod wykrywania początku i końca fragmentów mowy w wypowiadanych sekwencjach. Eksperymenty zostały przeprowadzone z użyciem środowiska Matlab i specjalnie utworzonej bazy krótkich wypowiedzi w języku polskim. Do budowy modeli mówców wykorzystano kwantyzacja wektorowa (VQ) oraz Gaussian Mixture Models (GMM). Podczas badań sprawdzono także wpływ obniżenia szybkości próbkowania na skuteczność identyfikacji mówcy.

10

Speaker Recognition System Based on GMM Multivariate Probability Distributions built-in a Digital Watermarking Token

84%

Lenarczyk P. , Piotrowski Z.

|

tom R. 89, nr 2a

59--63

PL

Przedstawiony poniżej artykuł opisuje system rozpoznawania mówcy na podstawie mowy ciągłej, wykorzystując wielowariancyjne rozkłady prawdopodobieństwa GMM. Opisane zostały procesy ekstrakcji cech dystynktywnych głosu oraz tworzenia modeli statystycznych. Algorytm został zaimplementowany w systemie Linux w celu poprawy funkcjonalności identyfikacji użytkownika Zaufanego Osobistego Terminalu PTT.

EN

The article describes a speaker recognition system based on continuous speech using GMM multivariate probability distributions. A theoretical model of the system including the extraction of distinctive features and statistical modeling is described. The efficiency of the system implemented in the Linux operating system was determined. The system is designed to support the functionality of the Personal Trusted Terminal PTT in order to uniquely identify a subscriber using the device.

11

Institutions and economic growth in Africa: Evidence from panel estimation

83%

Epaphra M. , Kombe A. H.

|

2017

|

tom Volume 13

|

nr Issue 5

570-590

EN

There is growing emphasis on the role of institutions on explaining Africa’s economic growth ahead of the traditional factors such as capital accumulation. However, it is not clear which of the institutions and governance indicators namely control of corruption, government effectiveness, political stability, regulatory quality, rule of law and voice and accountability matter most. This paper empirically examines the impact of institutions on economic growth in Africa. The paper uses a sample of 48 countries for the1996-2016 period. The overall number of observations is 912. The paper applied generalized methods of moment (GMM), fixed effects (FE) and random effects (RE) models. However, due to the fact that GMM is well suited to deal with potential endogeneity problems in the model, inferential statistics of this paper are drawn from GMM regression results. Results of the FE and RE regressions are presented in appendix. Empirical results show that institutions really matter for Africa’s economic growth. Among the institutional quality indicators political stability appears to be the most significant factor in explaining real GDP per capita growth in Africa. However, it is worth noting that, the quality of institutions alone may not be sufficient. Along with institutions, the paper reveals that structural factors such as liberalization of trade, fixed capital formation, labour force and foreign direct investment have a significant effect on Africa’s economic growth. The implication is that, a policy mix with the aim of improving the quality of institutions as well as reducing trade restrictions, enhancing both domestic and foreign investment and improving the quality of labour force would enhance economic growth in Africa.

12

Staff turnover and organizational performance: The case of a microfinance organization

83%

Giuliano R. , Godfroid C. , Radermecker L.

Journal of Economics and Management

|

2023

|

tom 45

413-440

EN

Aim/purpose – This study examines the relationship between staff turnover and performance in the microfinance industry in a dynamic perspective and investigates some contingency factors that moderate this relationship. Design/methodology/approach – We ran random-effects and GMM models based on a database of 2,814 branch-month observations from a specific microfinance organization. Findings – It takes three months to see a significant negative impact of turnover on the volume of a branch’s loan portfolio. Moreover, it takes four months after the turnover event for this negative impact to be counterbalanced. After four months, turnover stops having negative consequences and even becomes advantageous in terms of loan portfolio growth, but this positive effect lasts only one month. The effect of turnover thus appears to be particularly limited in time. Finally, we find that the negative relationship between turnover rate and performance is weakened by the seniority level of departing loan officers and by the recruitment rate. Originality/value/contribution – First, this paper examines the duration of the consequences of turnover event, which is poorly studied in the literature. Second, it focuses on microfinance, an industry where relational capital is of high importance. Third, it extends the theory on turnover by highlighting that the seniority level of departing employees is a moderator in the relationship between turnover and organizational performance.

13

NEW EMPIRICAL EVIDENCE ON THE DRIVERS OF FOREIGN DIRECT INVESTMENT IN THE EUROPEAN UNION COUNTRIES

83%

Donath L. , Miloş L. R. , Miloş M. C.

|

tom 68

|

nr 1

69 – 84

EN

For a number of decades, the European Union has witnessed an intense global competition to attract foreign direct investment (FDI). This competition has led to significant inflows and outflows in the member states, with positive implications for economic growth, employment and productivity for the host and sending countries. The paper uses data from 2005 – 2017 to identify the main drivers of FDI in the EU separately for sending and receiving, Euro area and Non-Euro countries. Analysis focuses on finding the levers that promote the Union in its entirety as a major competitor for FDI. An empirical analysis is conducted for selected variables (economic development, taxation, unit labour costs, trade openness, interest rate differential, macroeconomic stability, infrastructure and competitiveness) using the two-step System Generalised Method of Moments (GMM). The results are robust and consistent with the supporting literature.

14

EUROPEAN COHESION POLICY TO THE RESCUE? REVISING (COUNTER)-CYCLICAL EFFECTS IN THE EU-28 RECIPIENT COUNTRIES

71%

Stiblarova L.

Ekonomický časopis (Journal of Economics)

|

2022

|

tom 70

|

nr 9-10

681 – 702

EN

This paper empirically verifies recent efforts of EU policymakers and government authorities to apply the European Cohesion Policy (ECP) as a possible counter-cyclical instrument to boost economies in their economic downturns. Compared to limited country-specific studies, we allow for an endogeneity issue between the business cycles and the ECP payments and apply a system GMM estimation for the EU-28 recipient countries sample in the time period 2000 – 2018. Even though the overall ECP payments follow a pro-cyclical path, the model estimations for the sub-periods based on individual programming periods reveal a time-varying cyclical character of the ECP. Whereas the observed pro-cyclicality can be mainly contributed to the period 2000 – 2006, we find the conditional counter-cyclical effects in 2007 – 2013 and 2014 – 2018, when a lower price level and obeying the rule of law seem to have an extra counter-cyclical dimension, confirming a rationale of the convergence criteria and the stability and the growth pact.

15

Designing Model Based Classifiers by Emphasizing Soft Targets

71%

El Jelali S. , Lyhyaoui A. , Figueiras-Vidal A. R.

|

2009

|

tom Vol. 96, nr 4

419-433

EN

When training machine classifiers, to replace hard classification targets by emphasized soft versions of them helps to reduce the negative effects of using standard cost functions as approximations to misclassification rates. This emphasis has the same kind of effect as sample editing methods, that have proved to be effective for improving classifiers performance. In this paper, we explore the effectiveness of using emphasized soft targets with generative models, such as Gaussian MixtureModels (GMM), and Gaussian Processes (GP). The interest of using GMMis that they offer advantages such as an easy interpretation and straightforward possibilities to deal with missing values. With respect to GP, if we use soft targets, we do not need to resort to any complex approximation to get a Gaussian Process classifier and, simultaneously, we can obtain the advantages provided by the use of an emphasis. Simulation results support the usefulness of the proposed approach to get better performance and show a low sensitivity to design parameters selection.

16

An application of New Keynesian models to inflation in Croatia

71%

Domić A.

|

nr 2

6-13

EN

Background: The inflation dynamics of Croatia is studied in the paper, with the review of applicable marginal cost proxies for the hybrid New Keynesian Phillips Curve (NKPC), and estimation of three specifications of the hybrid NKPC for Croatia. Objectives: The goal of this research is to examine the effect of labor's share of income, the price of energy, and the price of imports and other open economy factors in driving inflation in Croatia from the first quarter of 2000 to the fourth quarter of 2011. Methods/Approach: We use the generalized method of moments (GMM) estimator to empirically estimate three NKPC specifications. The J-stat and Cragg-Donald F-test are used to test for overidentification and for weak instruments, respectively. Results: We find that the marginal cost proxy for the energy-augmented specification is statistically significant and quantitatively the largest, whereas those for the other two are statistically significant, but quantitatively negligible. Conclusions: The results provide an empirical contribution both to the literature on inflation in Croatia and the literature of the NKPC in a small open economy. We can conclude that the price of energy has been the strongest driver of inflation, whereas the open economy factors we tested have had very little influence.

17

Application of Gaussian mixture model and proteomic databases in the mass spectra analysis – architecture of software of comprehensive mass spectrometry data processing

71%

Plechawska M.

|

tom Vol. 31, nr 2A

289-302

EN

This paper presents assumptions of the protein mass spectra analysis software. All the spectra are modeled with Gaussian Mixture Models. Estimation of model parameters is done by means of an Expectation-Maximization algorithm. The obtained parameters are used for a further biological analysis. The software is integrated with four huge protein databases available on-line. The biological information about proteins may be achieved on the chosen level of some detail.

PL

Artykuł przedstawia założenia systemu służącego analizie widm masowych białek. Widma są modelowane za pomocą mieszanin rozkładów normalnych. Oszacowanie ich parametrów oparte jest na algorytmie Expectation-Maximization. Uzyskane parametry są poddawane dalszej analizie biologicznej. System jest zintegrowany z czterema dużymi białkowymi bazami danych dostępnymi on-line. Informacja biologiczna dotycząca zawartych w widmie białek może być uzyskana na wybranym stopniu szczegółowości.

18

Exchange rate volatility and financial performance of agriculture firms in Malaysia: An empirical analysis using GARCH, wavelet and system GMM

71%

Reaz M. , Mahat F. , Dahir A. M. , Sahabuddin M. , Al Mahi A. S. M. M.

|

2017

|

tom Volume 13

|

nr Issue 3

409-427

EN

The insurgence of exchange rate volatility over the years has gained the attention of not only scholars but also policy makers around the world. This paper investigates the influence of exchange rate volatility to the financial performance of agriculture firms in Malaysia. Authors use the system GMM dynamic panel techniques, wavelet coherence technique and GARCH (1, 1) for the period of 2001 and 2015. The findings show that the volatility of exchange rate of Malaysian Ringgit (RM) has a negative impact on the financial performance of agriculture firms in Malaysia. The ARME and AVA demonstrate a positive impact on the financial performance at 1% significance level for the full sample. The findings also reveal that financial performance, exchange rate, consumer price index, and interest rate comove while using the wavelet coherence.

19

Integracja polskiego rynku papierów udziałowych i rynku papierów dłużnych z rynkami zagranicznymi a wzrost gospodarczy w Polsce

67%

Bukowski S.

Annales Universitatis Mariae Curie-Skłodowska, sectio H – Oeconomia

|

2017

|

tom 51

|

nr 5

EN

The aim of the research, which is the basis of this paper, is to answer the question whether integration of Polish equity and debt securities markets with foreign market influence economic growth in Poland. The following hypothesis has been put forward: integration of Polish equity and debt securities market with foreign markets has a significant impact on the capital market development and economic growth in Poland. The results of the research with implementation of econometric models confirm the hypothesis.

PL

Celem artykułu jest odpowiedź na pytanie, czy integracja polskiego rynku udziałowych papierów wartościowych oraz rynku papierów dłużnych z rynkami zagranicznymi wpływa na wzrost gospodarczy w Polsce. Sformułowano hipotezę badawczą, że integracja polskiego rynku udziałowych papierów wartościowych oraz integracja polskiego rynku papierów dłużnych z zagranicą wpływa w sposób statystycznie i ekonomicznie istotny na rozwój rynku kapitałowego oraz wzrost gospodarczy w Polsce. Badania z zastosowaniem modeli ekonometrycznych wykazały, że wzrost stopnia integracji polskiego rynku udziałowych i dłużnych papierów wartościowych z zagranicą wywiera pozytywny i dość znaczny wpływ zarówno na rozwój rynku akcji, jak i na wzrost gospodarczy w Polsce.

20

Rozpoznawanie mówców

67%

Dustor A. , Izydorczyk J.

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2003

|

tom nr 2-3

71-76

PL

Omówiono problematykę identyfikacji i weryfikacji mówcy. Przedstawiono poszczególne elementy składowe systemu rozpoznawania mówcy skupiając się szczególnie na zagadnieniach ekstrakcji parametrów z sygnału mowy, tworzeniu modeli mówcy zarówno parametrycznych jak i nieparametrycznych oraz na metodach rozpoznawania. Skrótowo przedstawiono również zagadnienia związane z zasobami mowy.

EN

The article presents fundamentals of speaker recognition as well as some basic problems of this technology like feature extraction, model training and recognition in more detail. Additionally a short description of speech corpora is included.