Student projects can be started any time (including holidays) and can span over semester boundaries.

Contacts: Paul Primus and Florian Schmid

Our current thematic focus is on:

• Efficient Deep Learning for Audio Classification, opens an external URL in a new window
• Data-efficient Acoustic Scene Classification (e.g., with Few, opens an external URL in a new window or Zero-Shot Learning, opens an external URL in a new window)
• Learning Audio Concepts from Natural-Language Supervision, opens an external URL in a new window
• Combining Audio and Large Language Models, e.g., Listen, Think, and Understand, opens an external URL in a new window

Projects related to the DCASE, opens an external URL in a new window Challenge, including topics from precious challenges.

Remark: We are open for new proposals - if you are interested in Intelligent Audio Processing, feel free to contact us with your idea!