|
|
Text Dependent Speaker Verification
Fux, Jan ; Glembek, Ondřej (referee) ; Matějka, Pavel (advisor)
The goal of this Bachelor's thesis was to design text dependent speaker recognition system. There were few systems tested for MIT database. This database contains recordings of 0.46s average length. Best case for recognition is to use a combination of DTW system using posterior probability estimation (posteriograms) as an output of Phoneme recognizer and acoustic SID system based on iVectors and PLDA (Probabilistic Linear Component Analysis). Fusion with Neural network gives the best results (EER). These are 17.84% EER for women and 16.38% for men. It's 49.9% relative improvement for women and 54.2% for men against acoustic recognition alone.
|
|
|
Emotional States of Humans and their Determination using Speech Record Analysis
Lněnička, Jakub ; Míča, Ivan (referee) ; Smékal, Zdeněk (advisor)
The aim of the diploma project is to find a method through which it will be possibleto classify the selected emotion from speech. At the beginning of the work deals with the description of the human body and their voice-generating operation. Furthermore, the text deals with the problem of the human voice into digital form.Great attention is paid to the parameters of the speech signal with an emphasis on describing the symptoms to help the selected emotion. The work deals with therecognition of emotions and a description of some of them. The main part is finding the best methods to reduce symptoms of segmental and suprasegmental speech utterances. The results of success was achieved by comparing the classification of selected emotions when using multiple methods and compare their results. The most important criterion in assessing the results ofthe reduction parameters of the speech signal, based on previous research in this area.
|
|
|
Multiplatform Application for Speaker Verification
Görig, Jan ; Matějka, Pavel (referee) ; Glembek, Ondřej (advisor)
Bachelor thesis considers speaker recognition without knowledge of spoken message. There are described current feature extraction methods and their evaluation using Gaussian mixture model. The practical output of this work is application for visualization of the recognition process. Developed application is cross platform and it uses Qt and BSAPI libraries.
|
|
|
Acoustic Scene Classification from Speech
Dobrotka, Matúš ; Glembek, Ondřej (referee) ; Matějka, Pavel (advisor)
The topic of this thesis is an audio recording classification with 15 different acoustic scene classes that represent common scenes and places where people are situated on a regular basis. The thesis describes 2 approaches based on GMM and i-vectors and a fusion of the both approaches. The score of the best GMM system which was evaluated on the evaluation dataset of the DCASE Challenge is 60.4%. The best i-vector system's score is 68.4%. The fusion of the GMM system and the best i-vector system achieves score of 69.3%, which would lead to the 20th place in the all systems ranking of the DCASE 2017 Challenge (among 98 submitted systems from all over the world).
|
|
|
Automatic vocal-oriented recognition of human emotions
Houdek, Miroslav ; Přinosil, Jiří (referee) ; Atassi, Hicham (advisor)
This master thesis concerns with emotional states and gender recognition on the basis of speech signal analysis. We used various prosodic and cepstral features for the description of the speech signal. In the text we describe non-invasive methods for glottal pulses estimation. The described features of speech were implemented in MATLAB. For their classification we used the GMM classifier, which uses the Gaussian probability distribution for modeling a feature space. Furthermore, we constructed a system for recognition of emotional states of the speaker and a system for gender recognition from speech. We tested the success of created systems with several features on speech signal segments of various lengths and compared the results. In the last part we tested the influence of speaker and gender on the success of emotional states recognition.
|
|
|
Recognition of Poses and Gestures
Jiřík, Leoš ; Hradiš, Michal (referee) ; Zemčík, Pavel (advisor)
This thesis inquires the existing methods on the field of image recognition with regards to gesture recognition. Some methods have been chosen for deeper study and these are to be discussed later on. The second part goes in for the concenpt of an algorithm that would be able of robust gesture recognition based on data acquired within the AMI and M4 projects. A new ways to achieve precise information on participants position are suggested along with dynamic data processing approaches toward recognition. As an alternative, recognition using Gaussian Mixture Models and periodicity analysis are brought in. The gesture class in focus are speech supporting gestures. The last part demonstrates the results and discusses future work.
|
|
|
Paralinguistic signals recognition in spoken dialogs
Mašek, Jan ; Míča, Ivan (referee) ; Atassi, Hicham (advisor)
This document describes the three methods for the detection and classification of paralinguistic expressions such as laughing and crying from usual speech by analysis of the audio signal. The database of records was originally designed for this purpose. When analyzing everyday dialogs, music might be included, so the database was extended by four new classes as speech, music, singing with music and usual speech with background music. Feature extraction, feature reduction and classification are common steps in recognizing for all three methods. Difference of the methods is given by classification process in detail. One classification of all six classes at once is proposed in the first method called straight approach. In the second method called decision tree oriented approach we are using five intuitive sub classifiers in the tree structure and the final method uses for classification emotion coupling approach. The best features were reduced by feature evaluation using F-ratio and GMM classifiers were used for the each classification part.
|
|
|
The relationship between banking sector development and innovation activities in countries in the European Union from 2010 to 2020
Le, Doanh Tuan
This paper examines the influences of EU banking sector development on innovation activities with the empirical evidence taking from 24 selected EU countries in the period of 11 years from 2010 to 2020. The sample of data is collected as secondary data and inserted into the linear regression model under the form of a dynamic short panel data. The chosen research method is called the two-step system GMM. The findings reveal that the growth of an EU country’s banking system supports its innovation activities. Moreover, this research also extends the positive impacts of an EU country’s financial markets development on its innovative-related activities.
|
|
|
Foreign-owned banks and host economies
Fišerová, Tereza ; Teplý, Petr (advisor) ; Mejstřík, Michal (referee)
In the past two decades, significant changes have been shaping and transforming the banking sectors worldwide. Among these trends we find an intensive surge in foreign bank ownership which is especially remarkable in the countries of the Central, Eastern and South-Eastern European region. Using the sample of 17 countries and filtering out more than 140 domestically-operating foreign-owned banks, we examine the determinants of their performance in relation to host country conditions and home country banking sector performance over the period of seven years between 2005 and 2011. Due to the topic's currency, we additionally provide an insight into the link between sovereign debt and bank ownership. By means of system GMM model, or fixed effects model, we reveal that macroeconomic fundamentals of the host country affect the foreign-owned banks' performance but do not suffice in explaining it fully. Moreover, the depth of the current crisis as demonstrated in the home country impacts negatively on the host-country-operating foreign-owned banks. We did not find any convincing evidence of the host sovereign debt and bank ownership nature of relationship.
|
|
|
Bank's performance in low and negative interest rate environment
Hanzlík, Petr ; Teplý, Petr (advisor) ; Tripe, David (referee) ; Witzany, Jiří (referee) ; Tůma, Zdeněk (referee)
Dissertation thesis Banks' performance in low and negative interest rate environment Author: Mgr. Petr Hanzl'ık Abstract This dissertation consists of four empirical papers that focus on the performance of banks in the low or even negative interest rate environment characteristic for the decade after the global financial crisis of 2007-2009. The first paper focuses on the analysis of a re- lationship between the net interest margin (NIM) of EU banks and market interest rates in a low-interest rate environment while controlling for the impact of market concent- ration by examining a large sample of annual data on 629 banks from EU countries for the 2011-2016 period. The results show a positive concave relationship between NIM and short-term interest rates, deterioration of NIM for all types of banks and a higher market concentration leading to higher NIM. In the second paper, we examine the determinants of NIM of European and US banks in a zero lower bound (ZLB) situation while control- ling for institutional design factors, i.e. difference between capital-based and bank-based financial markets. We analyse a large sample of annual data on 629 European banks and 526 US during the 2011-2016 period confirming that NIM is significantly influenced by the different institutional designs. The third paper deals with...
|
guest ::
