Acoustic event classification and detection are important for many real-world applications such as ambient assisted living, security surveillance, meeting room transcription, human-computer interaction, and multimedia retrieval. In this work, the aim is to investigate novel machine learning techniques for acoustic event classification and detection. In particular, deep neural networks are of interest. More specific tasks are formulated in the DCASE challenge (http://dcase.community/)

We offer :

• existing code of CNNs

Your Tasks :

• develop neural network models in Tensorflow
• analyze the implemented systems in terms of accuracy and performance
• contribute to scientific work in form of a paper

Your Outcome :

• learn to implement Neural Networks on a GPU
• learn how to solve a difficult speech processing task
• get a broad education in applied machine learning

Your Profile :

• motivation and reliability are a prerequisite
• good knowledge in machine learning and neural networks (at least 2 machine learning courses)
• knowledge in python programming

Additional Information :

This thesis project is planned for a duration of 6 months starting immediately.

Contact :

Franz Pernkopf (pernkopf@tugraz.at)