Snoring classified: The Munich-Passau Snore Sound Corpus

Christoph Janott, Maximilian Schmitt, Yue Zhang, Kun Qian, Vedhas Pandit, Zixing Zhang, Clemens Heiser, Winfried Hohenhorst, Michael Herzog, Werner Hemmert, Björn W. Schuller

Abstract: Highlights•Snore sound excitation locations can be distinguished by acoustic properties.•Automatic classification models based on speech-features prove successful.•The ComParE feature set, used successfully in paralinguistics, showed best results.•Mel Frequency Cepstral Coefficients (MFCCs) were the best-performing single subset.•Formant-based features alone yielded inferior results.