• 강사 : 송지은 교수 (KAIST 인문사회과학부)
• 일시 : 2021. 12. 9 (목) 17:00~18:30

Speech understanding in real life can be challenging due to a variety of factors, such as background noise, characteristics of the speech signal (e.g., foreign and regional accents) and receiver limitations (e.g., non-native listeners). In this talk, I’ll explain how each of the environmental and listener- and speaker-related factors impairs speech intelligibility and what linguistic and cognitive mechanisms listeners use to cope with the difficulties. To investigate these issues in speech recognition, electroencephalography (EEG) can be used to measure brain responses to speech with high temporal precision. I will introduce my research using EEG, which has demonstrated that listeners exert greater cognitive effort and flexibly draw on linguistic information available in the sentence to overcome the demands of adverse listening conditions. I’ll also discuss how these findings on human listeners can help speech engineers to develop technologies that aid speech communication in everyday life.