8 papers
From Read Speech to Spoken Digits: A Task-Specific Evaluation of Speech Privacy With Informed Attackers
Jule Pohlhausen, Anjana Rajasekhar, Anna Leschanowsky +1
Protecting speech privacy in real-life audio recordings is a growing concern. This contribution evaluates the effectiveness of three obfuscation techniques in protecting linguistic…
Assessing the Impact of Noise and Speech Enhancement on the Intelligibility of Speech Codecs
Lyonel Behringer, Anna Leschanowsky, Anjana Rajasekhar +2
Preserving speech intelligibility is a minimum requirement for speech codecs in communication. Recently, very low-bitrate neural codecs have gained interest for replacing classical…
You Are What You Say: Exploiting Linguistic Content for VoicePrivacy Attacks
Ãnal Ege Gaznepoglu, Anna Leschanowsky, Ahmad Aloradi +4
Speaker anonymization systems hide the identity of speakers while preserving other information such as linguistic content and emotions. To evaluate their privacy benefits, attacks…
Benchmarking Neural Speech Codec Intelligibility with SITool
Anna Leschanowsky, Kishor Kayyar Lakshminarayana, Anjana Rajasekhar +4
Speech intelligibility assessment is essential for evaluating neural speech codecs, yet most evaluation efforts focus on overall quality rather than intelligibility. Only a few pub…
Transparent NLP: Using RAG and LLM Alignment for Privacy Q&A
Anna Leschanowsky, Zahra Kolagar, Erion Ãano +4
The transparency principle of the General Data Protection Regulation (GDPR) requires data processing information to be clear, precise, and accessible. While language models show pr…
Expert-Generated Privacy Q&A Dataset for Conversational AI and User Study Insights
Anna Leschanowsky, Farnaz Salamatjoo, Zahra Kolagar +1
Conversational assistants process personal data and must comply with data protection regulations that require providers to be transparent with users about how their data is handled…