Search Results

Now showing 1 - 3 of 3

Drum Accompaniment Generation Using Midi Music Database and Swquence To Sequence Neural Network
(Izmir Institute of Technology, 2022) Akyüz, Yavuz Batuhan; Gümüştekin, Şevket
This thesis aims to create an artificial intelligence model to reinterpret the drum parts of musical pieces and/or to accompany music with new uniquely generated drum patterns. Besides providing rhythmic indicators, drum parts are essential to emphasize emotions. Every instrument in a musical composition is in harmony with each other to be meaningful as a whole. Based on this observation, in this thesis, a MIDI dataset and an LSTM based Seq2Seq model were used to create a link between different instruments and drums. Before the training, we created a dataset involving midi pieces with drum parts and grouped them as input and output, which are non-drum instruments, and drum parts respectively. The model was trained with six different genres and the teacher forcing method was utilized to improve the training. After the training, at the generation stage, we made it possible to adjust the complexity of the generated drum parts by changing the temperature value, which we called the complexity value, using the temperature sampling method. We also created a user interface with an instrument selection pane to give users control over the drum instruments generated. Moreover, we proposed a novel approach to generalize the idea for not only MIDI data but also WAV data. To accomplish this task, Mel-spectrogram, MFCC, and tempogram features were used. Both proposed methods are shown to produce high-quality unique drum accompaniments for different genres with adjustable complexity and freedom of choosing the desired drum instruments.
Detection and Localization of Motorway Overhead Directional Signs by Convolutional Neural Networks Trained With Synthetic Images
(Izmir Institute of Technology, 2019) Hekimgil, Hakan; Baştanlar, Yalın
Image classification, object detection and recognition have gone a long way in the last decade. The competitions, starting with ImageNet, have shown that various improving implementations of Artificial Neural Networks are the best Machine Learning techniques at the time for such tasks. However, machine learning methods require much training data and the such data for image related tasks come at a cost in terms of time and effort, if it can be obtained at all. When training data is scarce or not representative of the whole target set, synthetic data and data augmentation methods are used to increase the training data using what is already available. This thesis work shows that when the target classification images have a structure, even a loose one, it is still possible to use machine learning methods, deep learning in this case, without any real data to begin with and still produce a good detection model. In this work, a Convolutional Neural Network model is trained to detect and localize informative motorway lane direction signs. Starting with no real samples of the target images, a large computer-generated training set is created to train the model. The resulting detector can detect the required sign types with high accuracy, localizing their position by bounding boxes and categorizing them.
A Systematic Evaluation of Semantic Representations in Natural Language Processing
(Izmir Institute of Technology, 2018) Sevgili Ergüven, Özge; Tekir, Selma
In the studies of semantics, the main aim is to address meaning. In a computational manner, this goal is accomplished through the encoding of language constructs. These encodings are in the form of information-theoretic measures and vector representations. We have focused on the representation of words. In word representations, the earlier approaches depend on counting the statistics between word and its accompanied words, whereas the current methods are based on learning approaches. At this point, we have investigated the relation between these two approaches. We have realized that both approaches use context as the normalization factor. We support our idea by evaluating word representations on some Natural Language Processing (NLP) tasks. Furthermore, we have studied the polysemous words which carry more than one meaning. The word representation of the polysemous word provides a representation that covers more than one meaning. To overcome this issue, we provide a method to create a representation for each sense of polysemous word.

Master Degree / Yüksek Lisans Tezleri

Browse

Filters

Settings

Sort By

Results per page

Search Results