Please ensure Javascript is enabled for purposes of website accessibility

Prof. Dr. Ümit Güz

Mühendislik ve Doğa Bilimleri Fakültesi

Elektrik - Elektronik Mühendisliği Öğretim Üyesi

Tam Zamanlı

Eğitim

2006-2007, Post-Doktora

Srı (stanford Research Instıtute) - International, Information And Computing Sciences Division (ıcs), Speech Technology And Research (star) Laboratory, Menlo Park, Ca, Usa. ., Srı (stanford Research Instıtute) - International, Information And Computing Sciences Division (ıcs), Speech Technology And Research (star) Laboratory, Menlo Park, Ca, Usa.


2007-2008, Post-Doktora

University Of California, Berkeley, International Computer Science Institute (ıcsı) Speech Group, Unıversıty Of Calıfornıa At Berkeley, Berkeley, Ca, Usa.


1988-1990, Önlisans

Yıldız Teknik Üniversitesi, Yıldız Meslek Yüksekokulu/ Bilgisayar Teknolojileri Bölümü/ Bilgisayar Programcılığı Pr.


1997-2002, Doktora

İstanbul Üniversitesi, Fen Bilimleri Enstitüsü/ Elektronik Mühendisliği (dr)


1995-1997, Yüksek Lisans

İstanbul Üniversitesi, Fen Bilimleri Enstitüsü/ Elektronik Mühendisliği (dr)


1990-1994, Lisans

İstanbul Üniversitesi, Mühendislik Fakültesi/ Elektrik-elektronik Mühendisliği Bölümü/ Elektronik Anabilim Dalı

Çalışma Alanları

Mühendislik Temel Alanı Elektrik-Elektronik ve Haberleşme Mühendisliği İşaret İşleme , Ses ve Konuşma İşleme , Makine Öğrenmesi

Ödüller

J. William Fulbright International Fellow

J. William FULBRIGHT Postdoc Scholar Program, AMERİKA BİRLEŞİK DEVLETLERİ

Yurtdışı Doktora Sonrası Araştırma Ödülü

TÜBİTAK BİDEB, TÜRKİYE

Akademik Makaleler

Bildiriler

Extension of Conventional Co-Training Learning Strategies to Three-View and Committee-Based Learning Strategies for Effective Automatic Sentence Segmentation

Modeling of EEG Signals by Using Predefined Functional Bases

Compression of the ElectroCardioGram ECG Signals by using Signature and Envelope Vector Banks

Modeling of Linear Time Invariant Systems by Using Frequency Domain Data

Modeling of the Linear Time Invariant Systems by using Frequency Domain Data

Modeling of the ECG ElectroCardioGram Signals with Unique Base Functions

Modeling of the Speech Signals with Base Sequences Generated from Different Sound Sources

A New Approach for Determination of the Optimum Signature and Envelope Functions

A New Approach for the Modeling of the EEG ElectroEncephaloGram Signals by using Optimum Base Functions

Modeling of the ElectroCardioGram ECG Signals with Generalized Signature and Envelope Vector Banks

Optimum Modelling of the Speech Signals with Generalized Speaker Independent Predefined Basic Sound and Envelope Vector Banks

Modeling of the ElectroMyoGram EMG Signals by using Unique Signature and Envelope Vector Banks

Definition of A New Isolated Word Recognition Algorithm by using Predefined Signature and Envelope Vector Banks

A Novel Language and Speaker Independent Method for the Modeling of the Speech Signals by using Signature and Envelope Function Sets

A Novel Electroencephalogram EEG Compression Technique

A Novel Approach on Compressing Medical Images SYMPES

Breast Cancer Data Modeling Imaged with the CT Modality via Predefined Energy and Pattern Blocks

A New Method to Represent Speech Signals via Predefined Functional Bases

A Novel Method to Represent ECG Signals via Predefined Personalized Signature and Envelope Functions

A New Modeling Method of the ElectroCardioGram ECG Signals Based on the Use of an Optimized Predefined Functional Database

A New Word Recognition Algorithm by Using Signature and Envelope Feature Spaces

A Novel Representation Method for ElectroMyoGram EMG Signals with Predefined Signature and Envelope Functional Bank

A Novel Method to Represent the Speech Signals by Using Language and Speaker Independent Predefined Functions Sets

An Efficient ECG Data Compression Technique Based on Predefined Signature and Envelope Vector Banks

A New Coding Method for Speech and Audio Signals

A New Speech Modeling Method SYMPES

Model Adaptation for Dialog Act Tagging

A Novel Fast Algorithm for Speech and Audio Coding

EEG Signal Compression Based on Classified Signature and Envelope Vector Sets

A New Algorithm for High Speed Speech and Audio Coding

Co training Using Prosodic and Lexical Information for Sentence Segmentation

A Novel Noise Robust and Low Bit Rate Speech Coding Algorithm

A Novel Hybrid Electrocardiogram Signal Compression Algorithm with Low Bit Rate

Compression of the CT Images using Classified Energy and Pattern Blocks

A Novel Image Compression Method Based on Classified Energy and Pattern Blocks Initial Results

A Novel Computed Tomography Image Compression Method based on Classified Energy and Pattern Blocks

A Novel Biometric Recognition Approach using Electrocardiogram Signals

A Novel Human Identification System based on Electrocardiogram Features

A New Speech Coding Algorithm using Zero Cross and Phoneme based SYMPES

Projeler

A New Image Compression Algorithm based on QB-UCEPB (Quadtree based Universally Classified Energy and Pattern Blocks) and DNN (Deep Neural Networks)

Elektrokardiyogram ve Konuşma İşareti Tabanlı Biyometrik Tanıma Sistemi

A New Low Bitrate Speech Coding Algorithm using Zero Cross and Phoneme Based SYMPES

Parmak Ucu EKG İşareti Tabanlı Biyometrik Tanıma Sistemi

Co training Using Prosodic Lexical and Morphological Information for Automatic Sentence Segmentation of Turkish Spoken Language

Extracting and Using Prosodic Information for Turkish Spoken Language Processing TUBITAK CAREER Project No 107E182

Natural Language Understanding in Human Human and Human Machine Spoken Dialogs Project No 05B304

Elektrokardiyogram (ECG) İşaretleri için Bir Matematiksel Model (Proje No 400 03062005)

A Novel ECG Modeling Technique Using Classified Signature and Envelope Vector Sets Project No 06B302

Automatic Sentence Segmentation with Semi Supervised Learning Algorithms Using Prosodic and Lexical Information Project No 09A301

Feature Extraction from Electrocardiogram ECG Signals for Biometric Recognition Project No 10A301

Image Compression using Classified Energy and Pattern Blocks Project No 10B301

Üyelikler

IEICE, Üye 2004

IEEE, Üye 2003

Dersler

Ders AdıDönemlerDili
ELEC1401 Logic Design2022-2023İngilizce
ELEC3504 Digital Signal Processing2022-2024İngilizce
ELEC3302 Electronics II2022-2024İngilizce
ELEC3301 Electronics I2022-2024İngilizce
ELEC2501 Signals and Systems2022-2024İngilizce
EE483 Digital Control Engineering2013-2014İngilizce
EE488 Special Topics (Applied Speech and Audio Signal Processing)2011-2014İngilizce
EE354 Digital Signal Processing2010-2014İngilizce
EE330 Electronics I2010-2013İngilizce
EE353 Signals and Systems2010-2014İngilizce
EE481 Special Topics (Applied Digital Signal Processing)2012-2014İngilizce
EE352 System Dynamics and Control2013-2014İngilizce
Ders AdıDönemlerDili
EE567 Two Dimensional Signal and Image Processing2022-2024İngilizce
EE550 Advanced Digital Signal Processing2022-2024İngilizce
EE586 Special Topics on Speech and Audio Signal Processing (Lisansüstü)2010-2011İngilizce

Yönetilen Tezler

YılHazırlayan Ad SoyadTez AdıÜniversite
2021ASENA EKŞİOĞLUA new model for speech compressionIşık Üniversitesi
2019RİFAT BERBERGİLBattery test and management systemIşık Üniversitesi
2018EGE MÜNİR SARIData compression for smart grid infrastructureIşık Üniversitesi
2017SEPIDEH NAHAVANDI GARGARICompression of the mammography images using quadtree based energy and pattern blocksIşık Üniversitesi
2017CEREN DURSUNBiometric recognition using bio-signalsIşık Üniversitesi
2017SALİH SİNAN TAYLANEnhancement of the coded speech using filteringIşık Üniversitesi
2016GÖKHAN GÜVENFingertip ECG signal based biometric recognition systemIşık Üniversitesi
2014İZEL D. REVİDİProsodic, morphological and lexical feature extraction of Turkish broadcast news dataIşık Üniversitesi
2013DOĞAN DALVAAutomatic speech recognition system for Turkish spoken languageIşık Üniversitesi
YılHazırlayan Ad SoyadTez AdıÜniversite
2022GÖKHAN GÜVENFingertip electrocardiogram and speech signal based biometric recognition systemIşık Üniversitesi
2018DOĞAN DALVACo-training using prosodic, lexical and morphological information for automatic sentence segmentation of Turkish spoken languageIşık Üniversitesi