Darko Pekar

29 papers A 1Misc 3Journal 5Unranked 20
YearRankTypeTitle / Venue / Authors
2025 — conf
Exploiting voice conversion in creating new TTS voices.
IWSSIP
Tijana V. Nosek, Sinisa Suzic, Nikola Simic, Milan Secujski, Darko Pekar, Vlado Delic
2025 — conf
Improving Whisper-Based Serbian ASR Using Synthetic Speech.
SPECOM (2)
Vuk Stanojev, Tijana V. Nosek, Sinisa Suzic, Darko Pekar, Vlado Delic, Milan Secujski
2024 — conf
End-to-End Speech Synthesis for the Serbian Language Based on Tacotron.
SPECOM (1)
Tijana V. Nosek, Sinisa Suzic, Milan Secujski, Vuk Stanojev, Darko Pekar, Vlado Delic
2024 — conf
Retrospective and Perspectives of TTS & STT Technology Development and Implementation for South Slavic Under-Resourced Languages.
SPECOM (1)
Milan Secujski, Branislav M. Popovic, Darko Pekar, Niksa Jakovljevic, Edvin Pakoci, Sinisa Suzic, Tijana V. Nosek, Nikola Simic, Vuk Stanojev, Vlado Delic
2022 — conf
HiFi-GAN based Text-to-Speech Synthesis in Serbian.
EUSIPCO
Sinisa Suzic, Darko Pekar, Milan Secujski, Tijana V. Nosek, Vlado Delic
2022 — conf
HiFi-GAN based Text-to-Speech Synthesis in Serbian.
EUSIPCO
Sinisa Suzic, Darko Pekar, Milan Secujski, Tijana V. Nosek, Vlado Delic
2022 — conf
Overcoming Data Sparsity in Automatic Transcription of Dictated Medical Findings.
EUSIPCO
Edvin Pakoci, Darko Pekar, Branislav M. Popovic, Milan Secujski, Vlado Delic
2021 J jnl
Cross-Lingual Neural Network Speech Synthesis Based on Multiple Embeddings.
Int. J. Interact. Multim. Artif. Intell.
Tijana V. Nosek, Sinisa Suzic, Darko Pekar, Radovan Obradovic, Milan Secujski, Vlado Delic
2021 — conf
Explicit Control of the Level of Expressiveness in DNN-Based Speech Synthesis by Embedding Interpolation.
SPECOM
Tijana V. Nosek, Sinisa Suzic, Mia Vujovic, Darko Pekar, Milan Secujski, Vlado Delic
2020 J jnl
Speaker/Style-Dependent Neural Network Speech Synthesis Based on Speaker/Style Embedding.
J. Univers. Comput. Sci.
Milan Secujski, Darko Pekar, Sinisa Suzic, Anton Smirnov, Tijana V. Nosek
2019 J jnl
Using Morphological Data in Language Modeling for Serbian Large Vocabulary Speech Recognition.
Comput. Intell. Neurosci.
Edvin Pakoci, Branislav M. Popovic, Darko Pekar
2018 — conf
A Comparison of Language Model Training Techniques in a Continuous Speech Recognition System for Serbian.
SPECOM
Branislav M. Popovic, Edvin Pakoci, Darko Pekar
2018 — conf
Toward More Expressive Speech Communication in Human-Robot Interaction.
ICR
Vlado Delic, Branislav Borovac, Milan Gnjatovic, Jovica Tasevski, Dragisa Miskovic, Darko Pekar, Milan Secujski
2017 — conf
End-to-End Large Vocabulary Speech Recognition for the Serbian Language.
SPECOM
Branislav M. Popovic, Edvin Pakoci, Darko Pekar
2017 — conf
Language Model Optimization for a Deep Neural Network Based Speech Recognition System for Serbian.
SPECOM
Edvin Pakoci, Branislav M. Popovic, Darko Pekar
2017 Misc conf
Novel alignment method for DNN TTS training using HMM synthesis models.
SISY
Sinisa Suzic, Tijana Delic, Darko Pekar, Vladimir Ostojic
2016 — conf
A Phonetic Segmentation Procedure Based on Hidden Markov Models.
SPECOM
Edvin Pakoci, Branislav M. Popovic, Niksa Jakovljevic, Darko Pekar, Fathy Yassa
2016 Misc conf
Advanced voice activity detection on mobile phones by using microphone array and phoneme-specific Gaussian mixture models.
SISY
Branislav M. Popovic, Edvin Pakoci, Darko Pekar
2016 — conf
An Algorithm for Phase Manipulation in a Speech Signal.
SPECOM
Darko Pekar, Sinisa Suzic, Robert Mak, Meir Friedlander, Milan Secujski
2014 — conf
Speaker Detection Using Phoneme Specific Hidden Markov Models.
SPECOM
Edvin Pakoci, Niksa Jakovljevic, Branislav M. Popovic, Dragisa Miskovic, Darko Pekar
2013 — conf
Speech and Language Resources within Speech Recognition and Synthesis Systems for Serbian and Kindred South Slavic Languages.
SPECOM
Vlado Delic, Milan Secujski, Niksa Jakovljevic, Darko Pekar, Dragisa Miskovic, Branislav M. Popovic, Stevan Ostrogonac, Milana Bojanic, Dragan Knezevic
2012 Misc conf
A language model for highly inflective non-agglutinative languages.
SISY
Stevan Ostrogonac, Dragisa Miskovic, Milan Secujski, Darko Pekar, Vlado Delic
2012 J jnl
A novel split-and-merge algorithm for hierarchical clustering of Gaussian mixture models.
Appl. Intell.
Branislav M. Popovic, Marko Janev, Darko Pekar, Niksa Jakovljevic, Milan Gnjatovic, Milan Secujski, Vlado Delic
2011 A conf
Automatic Prosody Generation for Serbo-Croatian Speech Synthesis Based on Regression Trees.
INTERSPEECH
Milan Secujski, Darko Pekar, Niksa Jakovljevic
2010 J jnl
Eigenvalues Driven Gaussian Selection in continuous speech recognition using HMMs with full covariance matrices.
Appl. Intell.
Marko Janev, Darko Pekar, Niksa Jakovljevic, Vlado Delic
2010 — conf
Naturalness, Adaptation and Cooperativeness in Spoken Dialogue Systems.
COST 2102 Training School
Milan Gnjatovic, Darko Pekar, Vlado Delic
2008 — conf
Energy Normalization in Automatic Speech Recognition.
TSD
Niksa Jakovljevic, Marko Janev, Darko Pekar, Dragisa Miskovic
2002 — conf
AlfaNum System for Speech Synthesis in Serbian Language.
TSD
Milan Secujski, Radovan Obradovic, Darko Pekar, Ljubomir Jovanov, Vlado Delic
1999 — conf
A robust speaker-independent CPU-based ASR system.
EUROSPEECH
Radovan Obradovic, Darko Pekar, Srdjan Krco, Vlado Delic, Vojin Senk