Branislav Gerazov
Professor at Faculty of Electrical Engineering and Information Technologies - Skopje
Skopje, Skopje Region, North Macedonia
Summary
👤
Senior🎓
Top SchoolBranislav Gerazov is a Professor and signal-processing researcher with over a decade of experience specializing in speech and audio processing, text-to-speech synthesis, intonation and prosody modelling, ASR, and speaker recognition. He combines academic leadership at the Faculty of Electrical Engineering and Information Technologies in Skopje with hands-on machine learning engineering—contributing code and training infrastructure improvements to the widely used coqui-ai/TTS deep learning toolkit. His work spans fundamental research (PhD in Technical Sciences) and applied biomedical signal processing and electronics, including deep-learning tumour classification and biomedical device coursework. A former visiting researcher at top European labs and a CTO for an audio startup, he brings both research rigor and product-oriented development to complex audio and biomedical problems. Notably, his contributions to prosody modelling in EU-funded projects demonstrate a rare blend of theoretical modelling and practical TTS system engineering.
10 years of coding experience
9 years of employment as a software developer
Doctor of Philosophy - PhD, Technical Sciences, Doctor of Philosophy - PhD, Technical Sciences at Ss. Cyril and Methodius University, SkopjeMaster of Engineering - MEng, Electronics, Master of Engineering - MEng, Electronics at Faculty of Electrical Engineering and Information Technologies - Skopje
Github Skills (9)
gtts10
freetts10
pytorch10
machine-learning10
text-to-speech10
deep-learning10
python10
model-optimization9
argument-parsing9
Programming languages (16)
Github contributions (5)
coqui-ai/TTS
Feb 2021 - Feb 2021
🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
Role in this project:
ML Engineer
ML EngineerContributions:12 reviews, 10 commits, 1 PR in 9 days
Contributions summary:Branislav made several commits focused on updating and modifying training scripts for various vocoder and TTS models within the coqui-ai/tts repository. Their changes include updating the training process for WaveGrad, GlowTTS, and other models, which involved modifications to data loading, loss computation, and model saving steps. Furthermore, they refactored argument parsing, indicating involvement in the core training infrastructure and configuration aspects of the project.
text-to-speechpythonspeaker-encodingsasrglow-tts
prosodylab/prosobeast-annotation-tool
Apr 2021 - Jan 2022
Contributions:1 release, 37 commits, 7 PRs in 9 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.