Assistant research scientist
Primary Appointment: Center for Language and Speech Processing
- Speech Recognition
- Spoken Language Processing
- Deep Learning
Piotr Zelasko is an assistant research scientist in the Center for Language and Speech Processing (CLSP) who is an expert on automatic speech recognition (ASR) and spoken language understanding (SLU).
His current research focuses on applying multilingual and cross-lingual speech recognition systems to categorize the phonetic inventory of a previously unknown language and on improving defenses against adversarial attacks on both speaker identification and automatic speech recognition systems. He is also addressing the question of how to structure a spontaneous conversation into high-level semantic units called dialog acts, and working on Lhotse + K2, the next-generation speech processing research software ecosystem.
Zelasko co-organized the Kaldi Community Forum – a virtual conference on the future directions of automatic speech recognition – during the summer of 2020. A member of Department of Electrical and Computer Engineering Associate Professor Najim Dehak’s group, Zelasko also mentors the group’s PhD and master’s students.
Before joining Johns Hopkins, Zelasko worked as a machine learning consultant for Avaya (2017-2019), and as a software engineer for Techmo (2015-2017). He is also a member of the International Speech Communication Association (ISCA), the Institute of Electrical and Electronics Engineers (IEEE) and IEEE’s Signal Processing Society (IEEE SPS).
Zelasko received his PhD (2019), as well as his master’s (2014) and undergraduate degrees (2013) in acoustic engineering from AGH University of Science and Technology in Kraków, Poland.