Our laboratory, together with Serbian colleagues, organized a special session "Multi-Lingual, Multi-Style, Multi-Modal Human-Machine Spoken Language Communication" at the 30th European Signal Processing Conference (EUSIPCO 2022). EUSIPCO is Europe's leading conference for signal processing.

Multi-Lingual, Multi-Style, Multi-Modal Human-Machine Spoken Language Communication Session chairs: Alexey Karpov, SPC RAS and Vlado Delić, University of Novi Sad Date: 1st September 2022 9 poster presentations from Serbia, Slovenia, Rep. Macedonia, Bosnia and Herzegovina, Hungary, Russia (almost all in-person)

Within the framework of this session, researcher Ivanko Denis presented results on the topic: “Visual Speech Recognition in a Driver Assistance System”, dedicated to automatic speech recognition based on video information. At the time of the presentation, in this work, the best results were obtained in automatic speech recognition by the speaker's lips, not only for Russian, but also for English speech. The co-authors of this work are also: Ryumin D., Kashevnik A., Aksenov A. and Karpov A.

Ivanko D., Ryumin D., Kashevnik A., Axyonov A., Karnov A. Visual Speech Recognition in a Driver Assistance System // In Proc. of 30th European Signal Processing Conference (EUSIPCO). 2022. pp. 1131-1135.