The INTERSPEECH 2017 Computational Paralinguistics Challenge: Addressee, Cold & Snoring

• DOI: 10.21437/interspeech.2017-43
• Publication Date: 2017-08-16T06:14:46Z
• Authors: Björn Schuller; Stefan Steidl; Anton Batliner; Elika Bergelson; Jarek Krajewski; Christoph Janott; Andrei Amatuni; Marisa Casillas; Amanda Seidl; Melanie Soderstrom; Anne S. Warlaumont; Guillermo Hidalgo; Sebastian Schnieder; Clemens Heiser; Winfried Hohenhorst; Michael Herzog; Maximilian Schmitt; Kun Qian; Yue Zhang; George Trigeorgis; Panagiotis Tzirakis; Stefanos Zafeiriou
• Keywords: ddc:004; 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

The INTERSPEECH 2017 Computational Paralinguistics Challenge addresses three different problems for the first time in research competition under well-defined conditions: In Addressee sub-challenge, it has to be determined whether speech produced by an adult is directed towards another or a child; Cold cold told apart from 'healthy' speech; and Snoring four types of snoring have classified.In this paper, we describe these sub-challenges, their conditions, baseline feature extraction classifiers, which include data-learnt representations end-to-end learning with convolutional recurrent neural networks, bag-of-audio-words challenge series.