Midia Yousefi

About

I have joined the Azure Cognitive Services Research team at Microsoft as a Senior Researcher. My research focuses on end-to-end meeting transcription with focus on speaker diarization and identification to improve the speaker-attributed ASR quality.

Prior to joining Microsoft in 2022, I was a PhD student at the University of Texas at Dallas. The research topic of my PhD dissertation was investigating deep-learning based methods for detection, separation, and recognition of overlapping speech.  During my PhD, I joined the Human-Machine Interaction (HMI) team at Robert Bosch LLC as an intern to work on target-speaker speech extraction in summer 2019. For my second internship, I joined the Audio and Acoustics group at Microsoft Research to work on audio-based toxic language detection for online gaming applications in summer 2020.