|
|
Reconstruction of signal modified by fade-in/fade-out
Bača, Petr ; Kiska, Tomáš (referee) ; Rajmic, Pavel (advisor)
This thesis contains the theory needed to solve the special problem of bit-depth expansion. The goal is to reconstruct the signal which suffered from application of the fade-in, fade-out effect. The theory includes information of analog to digital conversion and the theory of sparse representations. Thesis formulates the task of bit-depth expansion and advices the algorithm to solve it. Furthermore, the realization of the issue is discussed and the results are given.
|
|
|
Music genre recognition using Music information retrieval techniques
Zemánková, Šárka ; Zvončák, Vojtěch (referee) ; Kiska, Tomáš (advisor)
This diploma work deals with music genre recognition using the techniques of Music Information Retrieval. It contains a brief description of the principle of this research area and its subfield called Music Genre Recognition. The following chapter includes selection of the most suitable parameters for describing music genres. This work further characterizes machine learning methods used in this field of research. The next chapter deals with the descriptions of music datasets created for genre classification studies. Subsequently, there is a draft and evaluation of the system for music genre recognition. The last part of this work describes the results of partial parameter analysis, dependence of genre classification accuracy on the amount of parameters and contains a discussion on the causes of classification accurancy for the individual genres.
|
|
|
Platform for subjective evaluation of video-sequences
Srnec, Tomáš ; Kiska, Tomáš (referee) ; Číka, Petr (advisor)
This bachelor thesis is focused on subjective video quality assessment. Used modern codecs such as H.264, H.265, VP8 and VP9 are described in first chapter. In the next part of the thesis, four methods of the subjective video assessment are being called, according to Recommendation ITU-T P.910. The practical part includes encoding of three videos, into four resolutions, for four codecs. Output of the thesis is JavaFX application, capable of playing used videos for participants of test, who are making judgment. Their results are real-time sent to MySQL server and directly in application evaluated into bar charts. According to our results, the best codec is VP9, before codec H.265, H.264 and VP8.
|
|
|
An alternative JPEG coder/decoder
Jirák, Jakub ; Kiska, Tomáš (referee) ; Rajmic, Pavel (advisor)
The JPEG codec is currently the most widely used image format. This work deals with the design and implementation of an alternative JPEG codec using proximal algorithms in combination with the fixation of points from the original image to suppression of artifacts created in common JPEG coding. To solve the problem, the prox_TV and then the Douglas-Rachford algorithm were used, for which special functions using l_1-norm for image reconstruction were derived. The results of the proposed solution are very good because they can effectively suppress the artefacts created and the result corresponds to the image with a higher set qualitative factor. The proposed method achieves very good results for both simple images and photos, but in the case of large images (1024 × 1024 px) and larger, a large amount of computing time is required, so the method is more suitable for smaller images.
|
|
|
Research of dynamics features comparing audio records
Zemánková, Šárka ; Smékal, Zdeněk (referee) ; Kiska, Tomáš (advisor)
This work deals with the analysis of parameters related to the dynamics of sound recordings. It contains a brief description of the history of sound processing in analogue and digital form and the process of audio signal processing nowadays. The following chapter includes selection of the most suitable parameters for describing an audio recording, especially those describing the dynamics. This work further characterizes the methods used in similar researches in the world. There is also a system designed to calculate 43 dynamic parameters and the possibilities of their analysis are outlined as well. 35 different interpretations of one musical work were compared. Finally, the calculated parameters were drawn into scatter plots and evaluated using visual cluster analysis.
|
|
|
Acoustic analysis of gender-related patterns in Parkinson's disease
Herinek, Denis ; Kiska, Tomáš (referee) ; Galáž, Zoltán (advisor)
The bachelor's thesis is about acoustic analysis of gender-related patterns in Parkinson's disease by analysing speech task: reading passage. Parkinson's disease manifests in all subsystems involved in speech production (respiration, phonation, articulation and prosody). The aim of this thesis is familirization with symptoms of this disorder and speech parameters influenced by this disorder. Thesis contains preprocessing, parametrization of speech signal and statistic analysis of parameters. System of speech signal processing is created in MATLAB programming language.
|
|
|
Recognizing the historical period of interpretation based on the music signal parameterization
Král, Vítězslav ; Mucha, Ján (referee) ; Kiska, Tomáš (advisor)
The aim of this semestral work is to summarize the existing knowledge from the area of comparison of musical recordings and to implement an evaluation system for determining the period of creation using the music signal parameterization. In the first part of this work are describe representations which can music take. Next, there is a cross-section of parameters that can be extracted from music recordings provides information on the dynamics, tempo, color, or time development of the music’s recording. In the second part is described evaluation system and its individual sub-blocks. The input data for this evaluation system is a database of 56 sound recordings of the first movement of Beethoven’s 5th Symphony. The last chapter is dedicated to a summary of the achieved results.
|
|
|
Musical instruments recognition from audio records using Music information retrieval techniques
Kárník, Radoslav ; Mucha, Ján (referee) ; Kiska, Tomáš (advisor)
This paper discusses design and implementation of classifying system for recognition of musical instruments from audio records with use of Musical Information Retrieval techniques. In the first part, paper describes parameters used for instrument classification, calculation of said parameters from records and reduction of feature vector. Next part is devoted to tuning and implementation of various classifiers with focus on neural networks. These classifiers ar further tested on records from IRMAS dataset wchich contain 11 musical instruments playing solo or with other instruments. Results of classifiers tested on different parameters and different numbers of instruments are discussed in the last part.
|
|
|
Automatic tagging of musical compositions using machine learning methods
Semela, René ; Galáž, Zoltán (referee) ; Kiska, Tomáš (advisor)
One of the many challenges of machine learning are systems for automatic tagging of music, the complexity of this issue in particular. These systems can be practically used in the content analysis of music or the sorting of music libraries. This thesis deals with the design, training, testing, and evaluation of artificial neural network architectures for automatic tagging of music. In the beginning, attention is paid to the setting of the theoretical foundation of this field. In the practical part of this thesis, 8 architectures of neural networks are designed (4 fully convolutional and 4 convolutional recurrent). These architectures are then trained using the MagnaTagATune Dataset and mel spectrogram. After training, these architectures are tested and evaluated. The best results are achieved by the four-layer convolutional recurrent neural network (CRNN4) with the ROC-AUC = 0.9046 ± 0.0016. As the next step of the practical part of this thesis, a completely new Last.fm Dataset 2020 is created. This dataset uses Last.fm and Spotify API for data acquisition and contains 100 tags and 122877 tracks. The most successful architectures are then trained, tested, and evaluated on this new dataset. The best results on this dataset are achieved by the six-layer fully convolutional neural network (FCNN6) with the ROC-AUC = 0.8590 ± 0.0011. Finally, a simple application is introduced as a concluding point of this thesis. This application is designed for testing individual neural network architectures on a user-inserted audio file. Overall results of this thesis are similar to other papers on the same topic, but this thesis brings several new findings and innovations. In terms of innovations, a significant reduction in the complexity of individual neural network architectures is achieved while maintaining similar results.
|
|
|
Recognition of music cover versions using Music Information Retrieval techniques
Martinek, Václav ; Zvončák, Vojtěch (referee) ; Kiska, Tomáš (advisor)
This master’s thesis deals with designs and implementation of systems for music cover recognition. The introduction part is devoted to the calculation parameters from audio signal using Music Information Retrieval techniques. Subsequently, various forms of cover versions and musical aspects that cover versions share are defined. The thesis also deals in detail with the creation and distribution of a database of cover versions. Furthermore, the work presents methods and techniques for comparing and processing the calculated parameters. Attention is then paid to the OTI method, CSM calculation and methods dealing with parameter selection. The next part of the thesis is devoted to the design of systems for recognizing cover versions. Then there are compared systems already designed for recognizing cover versions. Furthermore, the thesis describes machine learning techniques and evaluation methods for evaluating the classification with a special emphasis on artificial neural networks. The last part of the thesis deals with the implementation of two systems in MATLAB and Python. These systems are then tested on the created database of cover versions.
|
guest ::
