Speaker change detection with privacy-preserving audio cues

In this paper we investigate a set of privacy-sensitive audio features for speaker change detection (SCD) in multiparty conversations. These features are based on three different principles: characterizing the excitation source information using linear prediction residual, characterizing subband spectral information shown to contain speaker information, and characterizing the general shape of the spectrum. Experiments show that the performance of the privacy-sensitive features is comparable or better than that of the state-of-theart full-band spectral-based features, namely, mel frequency cepstral coefficients, which suggests that socially acceptable ways of recording conversations in real-life is feasible. Categories and Subject Descriptors H.3.1 [Information Storage and Retrieval]: Content Analysis and Indexing General Terms Human Factors Keywords Modeling social interactions, Multiparty conversations, Speaker change detection, Privacy-sensitive features

• Type: Conference
• Year: 2009
• Where: ICMI
• Authors: Sree Hari Krishnan Parthasarathi, Mathew Magimai-Doss, Daniel Gatica-Perez, Hervé Bourlard
• Keywords: Biometrics, ICMI 2009, Multiparty Conversations, Privacy-sensitive Audio Features, Speaker Change Detection