Georgi Shopov

I am a Ph.D. student in Computer Science at the Institute of Information and Communication Technologies, Bulgarian Academy of Sciences, where I am supervised by Prof. D.Sc. Stoyan Mihov. My research focuses on leveraging automata theory to develop more expressive and computationally efficient language models by overcoming the fixed left-to-right generation order of traditional autoregressive models.

Research Interests

• Formal Languages and Automata Theory
• Data Structures and Algorithms
• Machine Learning

Teaching

I have served as a teaching assistant at the Faculty of Mathematics and Informatics, Sofia University for the following courses:

• Discrete Mathematics
• Applications of Finite Automata
• Languages, Automata and Computability
• Object-Oriented Programming
• Introduction to Programming

Theses

1. Finite Transducers for Representation of Language Models.
   Ph.D. Thesis, Institute of Information and Communication Technologies, Bulgarian Academy of Sciences, 2025.
2. Construction of Character-Level f-Transducers for Representation of Language Models.
   M.Sc. Thesis (in Bulgarian), Faculty of Mathematics and Informatics, Sofia University, 2019.

Publications

1. Georgi Shopov and Stefan Gerdjikov.
   Consistent Bidirectional Language Modelling: Expressive Power and Representational Conciseness.
   In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024.
2. Diana Geneva, Georgi Shopov and Stoyan Mihov.
   Algorithms for Probabilistic and Stochastic Subsequential Failure Transducers.
   In International Journal of Foundations of Computer Science, 34(8), 2023.
3. Diana Geneva, Georgi Shopov, Kostadin Garov, Maria Todorova, Stefan Gerdjikov and Stoyan Mihov.
   Accentor: An Explicit Lexical Stress Model for TTS Systems.
   In Proceedings of the 24th Annual Conference of the International Speech Communication Association, 2023.
4. Georgi Shopov, Stefan Gerdjikov and Stoyan Mihov.
   StreamSpeech: Low-Latency Neural Architecture for High-Quality On-Device Speech Synthesis.
   In Proceedings of the 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, 2023.
5. Diana Geneva, Georgi Shopov and Stoyan Mihov.
   Algorithms for Probabilistic and Stochastic Subsequential Failure Transducers.
   In Proceedings of the 25th International Conference on Implementation and Application of Automata, 2021.
6. Diana Geneva, Georgi Shopov and Stoyan Mihov.
   Composition and Weight Pushing of Monotonic Subsequential Failure Transducers Representing Probabilistic Models.
   In arXiv:2003.09364 [cs.FL], 2020.
7. Diana Geneva, Georgi Shopov and Stoyan Mihov.
   Building an ASR Corpus Based on Bulgarian Parliament Speeches.
   In Proceedings of the 7th International Conference on Statistical Language and Speech Processing, 2019.
8. Diana Geneva and Georgi Shopov.
   Towards Accurate Text Verbalization for ASR Based on Audio Alignment.
   In Proceedings of the 6th Student Research Workshop Associated with the 12th International Conference on Recent Advances in Natural Language Processing, 2019.