Ina Kodrasi heads the Signal Processing for Communication group at the Idiap Research Institute. She received her MSc in Communications, Systems and Electronics from Jacobs University Bremen and her PhD from the University of Oldenburg, where she also worked as a postdoctoral researcher on speech dereverberation and noise reduction. Her group develops signal processing and machine learning methods for the challenges that arise in human communication, considered from the perspective of the speaker (speaking impairments), the listener (hearing impairments), and the acoustic environment (noise and reverberation). Current work includes the automatic detection and characterisation of pathological speech, such as dysarthria, representation learning that separates speaker identity from clinical signal properties, and pathological speech enhancement. She is an elected member of the IEEE Technical Committee on Audio and Acoustic Signal Processing and has served repeatedly as Area Chair for ICASSP and Interspeech.