Postdoctoral and PhD Positions in Compressed Diffusion Models for Speech Processing at University of Hamburg

The University of Hamburg is offering exciting opportunities for both postdoctoral researchers and PhD students to join the BMBF-funded COMFORT project, focusing on 'Compressed Diffusion Models for Speech Processing.' The project aims to develop efficient, robust, and flexible AI methods, contributing to sustainable and trustworthy machine learning. Research will center on lightweight diffusion models for audio, compressed data representations, and model robustness under challenging conditions. The successful candidates will collaborate closely with partners at DESY and FAU, and will have opportunities for publications and conference presentations. The COMFORT project addresses the challenges of data-driven methods in AI, such as high hardware requirements, energy consumption, and model specialization. It seeks to create mathematically grounded methods for efficient, flexible, and robust learning, with a focus on model compression and transferability across various data types, including audio. The project is funded by the German Federal Ministry of Education and Research (BMBF) and runs from October 2024 to September 2027. Applicants should have a strong background in signal processing, machine learning, or related fields. For postdoctoral roles, a completed PhD is required; for PhD positions, a Master’s degree is necessary. Experience with audio or speech processing, diffusion models, or compressed data representations is highly desirable. Good programming skills and proficiency in English are expected. The positions are fully funded according to the E13/E14 salary scale, covering salary and research expenses for the duration of the project. To apply, visit the University of Hamburg Signal Processing Group job offer page for full details and submit your application by March 15, 2026. This is an excellent opportunity to contribute to cutting-edge research in signal processing, machine learning, and AI, with a focus on sustainability and robustness in speech and audio data analysis.