Egor Lakomkin

Papers

Total Citations

H-Index

About

Egor Lakomkin is a researcher specializing in affective computing, speech emotion recognition, and human-robot interaction, with a particular focus on applying deep learning techniques to enable more natural and effective communication between humans and machines. His work addresses critical challenges in making robots emotionally aware, a capability essential for safe and intuitive collaboration in real-world environments. Among his notable contributions, Lakomkin has investigated the robustness of speech emotion recognition (SER) systems built on deep neural networks, exploring their practical applicability beyond controlled laboratory settings — his most-cited work in this area has garnered 5 citations. He has also advanced multimodal emotion recognition by integrating end-to-end speech recognition models with sentiment analysis pipelines, demonstrating the importance of linguistic modality in affective state estimation. Additionally, his work on EmoRL introduced deep reinforcement learning as a novel framework for continuous acoustic emotion classification, offering robots real-time emotional awareness to detect potentially unsafe situations. Collectively, Lakomkin's research pushes the boundaries of emotionally intelligent systems, bridging gaps between machine learning theory and practical robotics applications, making him a noteworthy contributor to the growing field of affective human-robot interaction.