Authorship Identification and Interpretable Natural Language Processing
We have developed the AdHominem framework for reliable and interpretable authorship identification from text. This system has recently won the PAN@CLEF 2020 challenge.
Follow the links for further information:
https://pan.webis.de/clef20/pan20-web/author-identification.html
https://sechuman.ruhr-uni-bochum.de/team/ , http://www.eg.bucknell.edu/~rmn009/
Security and privacy of voice assistants
Personal assistants such as Alexa, Siri, or Cortana are widely deployed these days. Automatic Speech Recognition (ASR) systems can translate and even recognize spoken language and provide a written transcript of the spoken language. Recent advances in the fields of deep learning and big data analysis supported significant progress for ASR systems and have become almost as good at it as human listeners.
In this work, we demonstrate how an adversary can attack speech recognition systems by generating an audio file that is recognized as a specific audio content by a human listener, but as a certain, possibly totally different, text by an ASR system.
Follow the links for further information:
https://unacceptable-privacy.github.io/