Шановна СМС-спільното,Запрошуємо на четверте засідання наукового семінару факультету комп'ютерних наук KSE в 2-ому сезоні 2025 року.Тема: "How AI Models Speak".During this seminar, we will explore the challenges and techniques behind enabling AI models to generate and process audio.We will talk about:🔹Why audio interface is challenging;🔹How to enable transformers to generate audio;🔹Approaches to audio representation problem;🔹How it all combines in modern conversational and text-to-speech models.Спікер: Yevhenii Azarov — deep learning research engineer at Respeecher and PhD student at KNU.🗓 Коли: Чт, 27.11 о 18:00.📍 Де: KSE (вул. М. Шпака 3), classroom 3.05 & Zoom: https://us02web.zoom.us/j/87915504065?pwd=LbGDwipaC8U7xyCGZUTzFJdPbvNHJB.1🔗Реєстрація: https://forms.kse.ua/ssokseorgua/form/ThefourthmeetingoftheStudentMathSeminarinthenewsea/formperma/2wyf_B1QlRRznlaGEiZN3P-l1gB-YfmVrTqUjBTyo_MThis seminar is designed for students, researchers, and anyone interested in AI, natural language processing, and speech technologies.