The INTERSPEECH 2013 computational paralinguistics challenge: social signals, conflict, emotion, autism

By Björn Schuller, Stefan Steidl, Anton Batliner, Alessandro Vinciarelli, Klaus R. Scherer, Fabien Ringeval, Mohamed Chetouani, Felix Weninger, Florian Eyben, Erik Marchi and 5 more. 2013

805 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Human-Computer Interaction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Björn Schuller
2. Florian Eyben
3. Carlos Busso
4. Anton Batliner
5. Stefan Steidl
6. Shrikanth S. Narayanan
7. Laurence Y. Devillers
8. Erik Marchi

In Cognitive Neuroscience

1. Björn Schuller
2. Klaus R. Scherer
3. Petri Laukka
4. Carlos Busso
5. Shrikanth S. Narayanan
6. Emmanuel Vincent
7. Laurence Y. Devillers
8. Felix Weninger

In Human-Computer Interaction

1. Julia Hirschberg
2. Björn Schuller
3. Stefanos P. Zafeiriou
4. Shiliang Zhang
5. Wen Gao
6. Shiqing Zhang
7. Klaus R. Scherer
8. Judee K. Burgoon

Citations in the corpus

• Privacy-Preserving Adversarial Representation Learning in ASR: Reality or Illusion? (2019, background): “…speech processing algorithms can infer not only the spoken contents from the speech signal, but also the speaker’s identity [5], intention [6–9], gender [10,11], emotional state [12–14], pathological condition [15–17], personality [18,19] and cultural [20,21] attributes to a great extent.”
• Adieu features? End-to-end speech emotion recognition using a deep convolutional recurrent network (2016, methodology): “Nevertheless, the majority of these works make use of commonly hand-engineered features have been used as input features, such as Mel-Frequency Cepstral Coefficients (MFCC), Perceptual Linear Prediction (PLP) coefficients, and supra-segmental features such as those used in the series of ComParE [4] ”
• The INTERSPEECH 2016 Computational Paralinguistics Challenge: Deception, Sincerity & Native Language (2016, methodology): “The official baseline feature set is the same as has been used in the three previous editions of the INTERSPEECH ComParE challenges [26, 27, 18].”
• Speech Emotion Recognition Using Deep Convolutional Neural Network and Discriminant Temporal Pyramid Matching (2018, methodology): “[62] reports the best performance of by using the large AVEC-2013 feature set [65] on the EMO-DB dataset, and the large ComParE feature set [66] on the eNTERFACE05 dataset.”
• The Geneva Minimalistic Acoustic Parameter Set (GeMAPS) for Voice Research and Affective Computing (2016, background): “, [12]).”
• Social signal classification using deep blstm recurrent neural networks (2014, methodology): “Note that the official competition evaluation measure of the ComParE 2013 Social Signals Sub-Challenge was chosen to be the unweighted average area-under-the-curve (UAAUC) [26].”
• Recent developments in openSMILE, the munich open-source multimedia feature extractor (2013, methodology): “Details on the feature set can be found in [8, 9].”
• Building Naturalistic Emotionally Balanced Speech Corpus by Retrieving Emotional Speech from Existing Podcast Recordings (2019, positive): “1, following the setting used in previous studies [47].”
• The INTERSPEECH 2020 Computational Paralinguistics Challenge: Elderly Emotion, Breathing & Masks (2020, methodology): “It contains 6 373 static features resulting from the computation of functionals (statistics) over low-level descriptor (LLD) contours [22, 21].”