The Team

Branislav Gerazov is an associate professor at FEEIT, UKIM. He has over 15 years of experience in speech technologies, especially text-to-speech synthesis and automatic speech recognition. For his Masters (2011) he built a Macedonian TTS system from scratch, and for his PhD (2014) he worked on a Macedonian speech recognition system for voice control of smart devices. He has worked extensively on intonation and prosody modelling for the purpose of speech synthesis in English, French, German, Mandarin, and Vietnamese. He was awarded a Marie Sklodowska-Curie Individual Fellowship to work on prosody at GIPSA-lab, Grenoble-INP, France. He is always exited to tackle new problems in the field of speech technology!

Zoran Ivanovski is a professor at FEIT, UKIM, and head of the Institute of Electronics and the DIPteam. His expertise is in digital image processing and Deep Learning models, as well as speech synthesis and recognition, and intonation modelling. He was the PI for the FEEIT team in the SP2: SCOPES Project on Speech Prosody coordinated by Idiap, Martigny, Switzerland. The project focused on the design and development of a physiologically inspired intonation model. He has also worked on speaker identification and diarisation in a project financed by ReTell Ltd., Sunbury on Thames, England. As head of the DIPteam he has extensive experience in working with industry and delivering research results under a tight schedule.

Dimitar Tashkovski is a professor at FEIT, UKIM, where he is serving as Dean since 2013. He has extensive knowledge of the field of digital signal processing, and its application in various domains, including speech and audio. Within the scope of the SAM-TAM project he helped in the design and implementation of an ASR system specialised for the automatic transcription of Macedonian television media. The system is a Large Vocabulary Continuous Speech Recognition system that reaches state-of-the-art performance. He has also worked on other industry financed projects related to media monitoring. Taskovski has extensive organisational and managerial skills that he has honed through PI positions and active participation in many research projects.

Risto Chavdarov is a system engineer at FEEIT. He has more than 10 years of experience in information technologies, especially in cloud systems, web technologies and computer security. Within his career, he has worked on the design and implementation of various web systems. He has experience in the implementation of accessibility and integration of assistive technologies in software systems.

Daniela Janeva, BSc, is enrolled in the MT master program at FEEIT. Her work is focused on Speech Synthesis, specifically speech synthesis based on brain waves recorded through Electrocorticography (ECoG). This cutting-edge technology has the potential of allowing fluent speech to individuals with paralysis, as well as thought driven speech communication.

Filip Mitrovski, BSc, is enrolled in the MT master program at FEEIT. His work is focused on Automatic Speech Recognition. He was actively engaged in the design and implementation of the Macedonian ASR system for media transcription in the SAM-TAM project. He is also actively exploring modern end-to-end Deep Learning based approaches to speech recognition.