National Repository of Grey Literature 12 records found  1 - 10next  jump to record: Search took 0.00 seconds. 
Audio Classification with Deep Learning on Limited Data Sets
Harár, Pavol ; Platoš,, Jan (referee) ; Šimák, Boris (referee) ; Mekyska, Jiří (advisor)
Standardní postupy diagnózy dysfonie klinickým logopedem mají své nevýhody, především tu, že je tento proces velmi subjektivní. Nicméně v poslední době získala popularitu automatická objektivní analýza stavu mluvčího. Vědci úspěšně založili své metody na různých algoritmech strojového učení a ručně vytvořených příznacích. Tyto metody nejsou bohužel přímo škálovatelné na jiné poruchy hlasu, samotný proces tvorby příznaků je pracný a také náročný z hlediska financí a talentu. Na základě předchozích úspěchů může přístup založený na hlubokém učení pomoci překlenout některé problémy se škálovatelností a generalizací, nicméně překážkou je omezené množství trénovacích dat. Jedná se o společný jmenovatel téměř ve všech systémech pro automatizovanou analýzu medicínských dat. Hlavním cílem této práce je výzkum nových přístupů prediktivního modelování založeného na hlubokém učení využívající omezené sady zvukových dat, se zaměřením zejména na hodnocení patologických hlasů. Tato práce je první, která experimentuje s hlubokým učením v této oblasti, a to na dosud největší kombinované databázi dysfonických hlasů, která byla v rámci této práce vytvořena. Předkládá důkladný průzkum veřejně dostupných zdrojů dat a identifikuje jejich limitace. Popisuje návrh nových časově-frekvenčních reprezentací založených na Gaborově transformaci a představuje novou třídu chybových funkcí, které přinášejí reprezentace výstupů prospěšné pro učení. V numerických experimentech demonstruje zlepšení výkonu konvolučních neuronových sítí trénovaných na omezených zvukových datových sadách pomocí tzv. "augmented target loss function" a navržených časově-frekvenčních reprezentací "Gabor" a "Mel scattering".
Front-end frameworks for ASP.NET Core web applications
Kozhukhov, Konstantin ; Harár, Pavol (referee) ; Lattenberg, Ivo (advisor)
In this bachelor thesis I deal with design and implementation of a web application servingas information system for storage of electronic components. The first two chapters ofthis work deal with the theoretical analysis and description of technologies that will beused in the application, while the last chapter is dedicated to implementation. The goalof the resulting program is a simple demonstration of the use of the platform ASP.NETCore with a framework DotVVM.
Audio noise reduction using deep neural networks
Talár, Ondřej ; Galáž, Zoltán (referee) ; Harár, Pavol (advisor)
The thesis focuses on the use of deep recurrent neural network, architecture Long Short-Term Memory for robust denoising of audio signal. LSTM is currently very attractive due to its characteristics to remember previous weights, or edit them not only according to the used algorithms, but also by examining changes in neighboring cells. The work describes the selection of the initial dataset and used noise along with the creation of optimal test data. For network training, the KERAS framework for Python is selected. Candidate networks for possible solutions are explored and described, followed by several experiments to determine the true behavior of the neural network.
Tool for Automatic Information Obtaning from the Web
Poliak, Jakub ; Harár, Pavol (referee) ; Povoda, Lukáš (advisor)
This bachelor thesis deals with programming of a tool for collecting positive and negative comments from one of the most popular Chinese e-shop to a database. It will be used for deep learning of an artificial neural network which should distinguish positive text from negative. Application was programmed in Java with the use of JSON-simple and jsoup libraries.
Creating a database of audio recordings with artificial noise in an anechoic chamber
Hájek, Vojtěch ; Povoda, Lukáš (referee) ; Harár, Pavol (advisor)
This bachelor thesis deals with theory of creating the database of sound records and subsequent creating the database of speech records in the anechoic chamber. Database was created as training dataset for learning process of the artificial neural network, which will be able to separate the speech from background noise. Therefore as the part of the database there are also the recordings of various types of noise that will be used as background noise for the voice recordings. The dataset contains records taken from 18 speakers aged from 16 to 76 years. Half of the speakers were men, half women. Database contains 405 records of speach of average length 46,7 secons and total length 315 minutes. By combining each speech record with each noise record at three levels of signal-to-noise ratio was created 7290 mixed records.
Semantic Recognition of Comments on the Web
Stříteský, Radek ; Harár, Pavol (referee) ; Povoda, Lukáš (advisor)
The main goal of this paper is the identification of comments on internet websites. The theoretical part is focused on artificial intelligence, mainly classifiers are described there. The practical part deals with creation of training database, which is formed by using generators of features. A generated feature might be for example a title of the HTML element where the comment is. The training database is created by input of classifiers. The result of this paper is testing classifiers in the RapidMiner program.
Deep Learning for Text Classification
Kolařík, Martin ; Harár, Pavol (referee) ; Povoda, Lukáš (advisor)
Thesis focuses on analysis of contemporary machine learning methods used for text classification based on emotion and testing several deep neural nework architectures. Outcome of this thesis is a neural network architecture, which is tuned for using with text data and which had the best result of 79,94 percent. Proposed method is language independent and it doesn’t require as precisely classified training datasets as current methods. Training and testing datasets were consisted of short amateur movie reviews in Czech and in English. Thesis contains also overview of theoretical basics for convolutional neural networks and history of neural networks and language processing Scripts were written in Python, neural networks were simulated using Keras library and Theano framework. We used CUDA for better performance.
Audio noise reduction using deep neural networks
Talár, Ondřej ; Galáž, Zoltán (referee) ; Harár, Pavol (advisor)
The thesis focuses on the use of deep recurrent neural network, architecture Long Short-Term Memory for robust denoising of audio signal. LSTM is currently very attractive due to its characteristics to remember previous weights, or edit them not only according to the used algorithms, but also by examining changes in neighboring cells. The work describes the selection of the initial dataset and used noise along with the creation of optimal test data. For creation of the training network is selected KERAS framework for Python and are explored and discussed possible candidates for viable solutions.
Tool for deep neural network design
Hladík, Jakub ; Harár, Pavol (referee) ; Uher, Václav (advisor)
The aim of this thesis was to create a program for visualization of artificial neural networks. The first part contains definition of artificial neural networks, comparison with biological neural networks and comparison with processors used in personal computers. Also contains brief description of their function and advantages/disadvantages of artificial neural networks. The second part contains brief explanation of deep learning. There are described methods and models used for learning. The third part contains introduction to Caffe framework and summary of the most frequently occuring layers in neural networks, that are created by using Caffe. Next part contains brief analysis of created program as well as detailed description of classes, that program contains.
The algorithm for the detection of positive and negative text
Musil, David ; Harár, Pavol (referee) ; Povoda, Lukáš (advisor)
As information and communication technology develops swiftly, amount of information produced by various sources grows as well. Sorting and obtaining knowledge from this data requires significant effort which is not ensured easily by a human, meaning machine processing is taking place. Acquiring emotion from text data is an interesting area of research and it’s going through considerable expansion while being used widely. Purpose of this thesis is to create a system for positive and negative emotion detection from text along with evaluation of its performance. System was created with Java programming language and it allows training with use of large amount of data (known as Big Data), exploiting Spark library. Thesis describes structure and handling text from database used as source of input data. Classificator model was created with use of Support Vector Machines and optimized by the n-grams method.

National Repository of Grey Literature : 12 records found   1 - 10next  jump to record:
Interested in being notified about new results for this query?
Subscribe to the RSS feed.