Sharif Digital Repository

In this thesis, we investigate speaker verification as a biometric technology to verify a person based on his/her claim. Text-dependent speaker verification systems are preferred in commercial and security applications and these systems have better performance in limited data condition based on a prior knowledge about speakers that are assumed to be cooperative. Limited amount of enrollment data is a major concern in this thesis. Speaker dependent model construction and channel variability issues on telephone-based text-dependent speaker verification applications are surveyed. Due to the lack of an appropriate database for the task, we collected a database which is referred to as text-prompt... 

Transformers are the key component in power system transmission and distribution networks. Condition based maintenance will increase their expected life and online monitoring is essential to ensure operation reliability. In this work a new approach to transformer online monitoring is provided based on partial discharge (PD) measurement.Multi-source PD signal separated using time-frequency S transform (ST) that is applied to the PD signal waveforms. The resultant ST matrix is then converted to gray scale image from which high level features are extracted using Bag of Words (BoW). Gaussian mixture model (GMM) clustering is used to discover clusters in the feature space. For recognition of... 

Existence of heterogeneity in vital tissues of complex multicellular organisms like mammals, and fatal tissues like cancer on one hand, and limited access to biological properties of their components on the other hand, turn the study of these tissue traits to one of the most interesting fields in bioinformatics. One of the hottest subjects in this field is the recognition of functional components of these tissues by using bulk data extracted from the whole tissue.Almost every method that aims to achieve such a purpose, particularly using gene expression data, assumes that all of the cell types which constitute the studied tissue have a deterministic expression profile.In this thesis we... 

In this paper, we introduce a new dataset for SAD and evaluate certain common methods such as GMM, ANN, and RNN on it. We have collected our dataset in a semi-supervised approach, using subtitled movies, with a labeling accuracy of 95%. This semi-automatic method can help us collect huge amounts of labeled audio data with very high diversity in language, speaker, and channel. We model the problem of SAD as a classification task to two classes of speech and non-speech. When using GMM for this problem, we use two separate mixtures to model speech and non-speech. In the case of neural networks, we use a softmax layer at the end of the network, with two neurons which represent speech and... 

The human speech conveys much information such as semantic contents, emotion and even speaker identity. Our goal in this thesis is the task of text-independent speaker identification (SI) in large population applications. Identification (test) time has become one of the most important issues in recent real time systems. Identification time depends on the cost of likelihood computation between test features and registered speaker models. For real time application of SI, system must identify an unknown speaker quickly. Hence the conventional SI methods cannot be used. The main goal in this thesis is to propose several methods that reduced identification time without any loss of identification... 

Nowadays, video search and retrieval is interesting for computer users and it has chief usages for multimedia systems. Video generation rate has increased and Internet as a communication framework is case of its transferring on the world. Because of these, importance of video files is more than past. Searching for finding content will be faster if video files would have indexed with a comprehensive system. The biggest step in this way is power of index generation that would be same or similar to human mind, for improvement of the clustering’s result or classification’s result. For generating suitable indexes, it is necessary to extracting effective features from videos and synthesizing these... 

The usage of robots as our friends has been proliferated these days. Knowing that they are going to be used in ordinary houses, we should develop methods and algorithms in order to provide a situation for end-users to program their own robots for their desired tasks. Learning from Demonstrations (LfD) can play a crucial role in this field. In this study, we had taught a non-verbal communication method (pointing) to a robot utilizing LfD. The learning method used was TP-GMM1. The rationale to use this method was that it models all the degrees of freedom together, and we thought it might be an essential parameter to make a movement more natural and understandable which could be two vital... 

Speech Recognition is an important topic in speech processing. In this thesis, we intend to do Isolated Word Recognition (IWR) a large vocabulary dataset. Previous works on large vocabulary IWR have used Hidden Markov Models, Gaussian Mixture Model and hybrid methods for this purpose, But our approach is based on Deep Neural Network (DNN). DNNs have shown excellent performance recently in different applications of voice and image processing. A key factor in speech recognition is the availability at appropriate datasets. There has been no acceptable speech corpus in Persian language for isolated word recognition before this work. In addition, Persian IWR systems reported so far are quite... 

Partial discharges that occur in a transformer insulation, generate current pulses. If these pulses be recorded, they can be used for transformer insulation condition assessment. Through processing of these recorded partial discharge signals, the PRPD patterns are generated and used to identify the source type of partial discharge defect. If multiple partial discharge defects exist in a transformer insulation, the related PRPD pattern, doesn’t look like any PRPD patterns of single defects. In this case, we need in the first step to discriminate the partial discharge signals stemmed from all the existing multiple partial discharge sources. To simulate the occurrence of multiple partial... 

A Brain–Computer Interface is a system that provides a direct pathway for communication between a brain and a computer device by processing signals from sensors measuring brain activity (here Electroencephalography signals). Brain signals are known to be stochastic, non-stationary, non-linear and highly noisy, Therfore Brain–Computer Interface Systems rely on signal preprocessing, feature extraction and use of machine learning methods in order to detect mental state of Brain–Computer Interface user. Current approaches addressing the problem are mainly based on supervised learning methods. In this Thesis, first some of freely obtainable datasets with motor or motor-imagery paradigms are...