PhD in Automatic Speech Processing and Speaker Diarization at University of Zurich University of Zurich in Switzerland

PhD opening in automatic speech processing and speaker diarization at the University of Zurich, Department of Computational Linguistics, Language Technology Group.

The position is part of the SAGE project (Speaker Recognition across Age-Groups for Cantonal Law Enforcement Agencies), funded by the Digital Society Initiative at UZH. The project addresses forensic speech technology challenges, including speaker number estimation, robust speaker embeddings, and diarization in difficult audio conditions such as children’s and female voices, poor audio quality, and multi-speaker conversations.

The doctoral researcher will work with Srikanth Madikeri, Volker Dellwo, and Thilo Stadelmann. The role is based at the Department of Computational Linguistics, UZH, and is described as a 80% temporary position starting in August 2026 or by agreement.

Eligibility highlights include a Master’s degree (or imminent Master’s degree) in computational linguistics, computer science, AI, or a related field; proficiency in Python; experience with machine learning toolkits; a strong academic record in machine learning and speech processing; and strong communication skills.

Funding is provided for three years, with remuneration according to SNSF salary rates (CHF 50,504 in year 1, CHF 52,115 in year 2, CHF 53,725 in year 3).

Application deadline: 2026-04-21. Applicants should submit a motivation letter, CV with publications, and two references via the UZH application portal.